{
 "title": "Hayz Labs reviews: scores of AI models, apps and tools",
 "license": "CC BY 4.0",
 "license_url": "https://creativecommons.org/licenses/by/4.0/",
 "credit": "Credit Hayz Labs (hayzlabs.com/reviews) when you reuse this. Arena leaderboard data © Arena, CC BY 4.0 (huggingface.co/datasets/lmarena-ai/leaderboard-dataset), and Epoch AI benchmark data, CC BY 4.0 (epoch.ai/benchmarks); our scores change and combine them by our own method.",
 "license_note": "Our scores, verdicts and write-ups are ours to share under CC BY 4.0: credit Hayz Labs. Values from others are not ours to license. Our model scores are built from public data, including leaderboards Arena publishes as an open dataset under CC BY 4.0 (credit \"Arena leaderboard data © Arena, CC BY 4.0\") and Epoch AI's benchmark results under CC BY 4.0 (credit \"Epoch AI\"). The 0 to 100 scores are our changes to them. Prices and specs are makers' published facts, linked on each row. We leave Artificial Analysis figures out of these files because its terms do not allow bulk reuse; review pages cite them with a link.",
 "third_party": [
  {
   "source": "Arena (arena.ai)",
   "fields": "an input to grades[] score, score_low and score_high where score_inputs names Arena; arena_rank; list.csv arena_places",
   "licence": "CC BY 4.0",
   "credit": "Arena leaderboard data © Arena, CC BY 4.0",
   "via": "https://huggingface.co/datasets/lmarena-ai/leaderboard-dataset",
   "changes": "the 0 to 100 scores and distinct-model counts are ours"
  },
  {
   "source": "Epoch AI (epoch.ai)",
   "fields": "an input to grades[] score, score_low and score_high where score_inputs names Epoch AI",
   "licence": "CC BY 4.0",
   "credit": "Epoch AI",
   "via": "https://epoch.ai/benchmarks",
   "changes": "the 0 to 100 scores are ours"
  },
  {
   "source": "Makers' own pages",
   "fields": "prices, plans, context sizes, licences",
   "licence": "facts, linked per row; not ours to license"
  },
  {
   "source": "Artificial Analysis (artificialanalysis.ai)",
   "fields": "none: its figures are left out of this file, and no score is worked out from them",
   "licence": "its terms forbid bulk or machine-readable copies"
  }
 ],
 "method": "https://hayzlabs.com/how-we-score/",
 "note": "Each job has a score from 0 to 100, by our own method. score_source \"model\" means it is built from public data, named in score_inputs; \"checklist\" means our own checklist. score_low and score_high are the likely range. A null score means no score yet (arena_rank alone is a rank in Arena's blind votes for agents). We never run or test anything ourselves. Our scores are built from public data, including Arena's open leaderboard dataset and Epoch AI's benchmark results, both CC BY 4.0, which we change and combine by our own method.",
 "arena_dataset": {
  "url": "https://huggingface.co/datasets/lmarena-ai/leaderboard-dataset",
  "license": "CC BY 4.0",
  "read": "2026-10-20"
 },
 "epoch_data": {
  "url": "https://epoch.ai/benchmarks",
  "license": "CC BY 4.0",
  "read": "2026-10-21T03:06:59.632Z"
 },
 "count": 63,
 "reviews": [
  {
   "slug": "deepseek-app",
   "url": "https://hayzlabs.com/reviews/deepseek-app/",
   "name": "DeepSeek app",
   "maker": "DeepSeek",
   "kind": "app",
   "license": "proprietary",
   "runs": "cloud",
   "released": "2026-10-17",
   "status": "first-look",
   "verdict": "Wait",
   "verdict_line": "Free, on a new model that Arena's open data doesn't rank yet. Your data is stored in China.",
   "grades": [],
   "price": {
    "free_plan": true,
    "usd_month": 0,
    "plans": "Free; no paid plan or in-app purchases listed (s6)",
    "api": null,
    "checked_on": "2026-10-18"
   },
   "how_we_rated": "First look from DeepSeek's release note and privacy policy and the App Store listing, checked 18 Oct 2026. Not rated yet: Arena's open leaderboard dataset does not list V4.1-Flash-1017 (its latest WebDev file was published 11 Sep), and our rule counts tests of an older version as no evidence. We did not run it ourselves. Where we point to an Anthropic product, we also name the best non-Anthropic option.",
   "sources": [
    {
     "id": "s1",
     "title": "Arena open leaderboard dataset: webdev, latest split (published 2026-09-11), CC BY 4.0",
     "url": "https://huggingface.co/datasets/lmarena-ai/leaderboard-dataset",
     "checked": "2026-10-18"
    },
    {
     "id": "s4",
     "title": "DeepSeek: DeepSeek-V4.1-Flash-1017 release (17 Oct 2026)",
     "url": "https://api-docs.deepseek.com/news/news261017/",
     "checked": "2026-10-18"
    },
    {
     "id": "s5",
     "title": "DeepSeek Privacy Policy (last updated 10 Feb 2026)",
     "url": "https://cdn.deepseek.com/policies/en-US/deepseek-privacy-policy.html",
     "checked": "2026-10-18"
    },
    {
     "id": "s6",
     "title": "Apple App Store: DeepSeek AI Assistant",
     "url": "https://apps.apple.com/us/app/deepseek-ai-assistant/id6737597349",
     "checked": "2026-10-18"
    }
   ],
   "checked": "2026-10-18"
  },
  {
   "slug": "grok",
   "url": "https://hayzlabs.com/reviews/grok/",
   "name": "Grok",
   "maker": "xAI",
   "kind": "app",
   "license": "proprietary",
   "runs": "cloud",
   "released": "2026-10-11",
   "status": "first-look",
   "verdict": "Worth trying",
   "verdict_line": "xAI's assistant in its own app and on X, free or from $8/month. No score yet: xAI doesn't say which model runs.",
   "grades": [],
   "price": {
    "free_plan": true,
    "usd_month": 0,
    "plans": "Free: 'Limited Chat requests'. Plus $8/month or $80/year: 'Grok Chat, Voice, Imagine, and Build', base usage. Premium $30/month or $300/year: 'Latest Grok models', 4x the usage of Plus. Super $100/month or $1,000/year: 15x, with 'Priority model access at peak times'. Ultra $200/month or $2,000/year: 50x. Teams from $30/user/month (s1, s2).",
    "api": null,
    "checked_on": "2026-10-18"
   },
   "how_we_rated": "First look from xAI's pricing page, plan docs, consumer FAQ and Grok 4.7 post, and X's Grok help page, checked 18 Oct 2026. Not rated yet. We did not run it ourselves.",
   "sources": [
    {
     "id": "s1",
     "title": "xAI pricing (Grok plans)",
     "url": "https://x.ai/pricing?v=1018e",
     "checked": "2026-10-18"
    },
    {
     "id": "s2",
     "title": "xAI docs: subscription pricing",
     "url": "https://docs.x.ai/x-subscription/pricing?v=1018e",
     "checked": "2026-10-18"
    },
    {
     "id": "s3",
     "title": "xAI Consumer FAQs",
     "url": "https://x.ai/legal/faq?v=1018e",
     "checked": "2026-10-18"
    },
    {
     "id": "s4",
     "title": "X Help Center: About Grok",
     "url": "https://help.x.com/en/using-x/about-grok?v=1018e",
     "checked": "2026-10-18"
    },
    {
     "id": "s5",
     "title": "xAI: Grok 4.7",
     "url": "https://x.ai/news/grok-4-7?v=1018e",
     "checked": "2026-10-18"
    }
   ],
   "checked": "2026-10-18"
  },
  {
   "slug": "opencode",
   "url": "https://hayzlabs.com/reviews/opencode/",
   "name": "OpenCode",
   "maker": "Anomaly",
   "kind": "open-source",
   "license": "open-source",
   "runs": "both",
   "released": "2026-10-11",
   "status": "graded",
   "verdict": "Worth trying",
   "verdict_line": "Free open-source coding agent for 75+ model providers; scored 54 with GLM-5.3. Skip it for one flat monthly plan.",
   "grades": [
    {
     "category": "coding",
     "score": 80,
     "score_low": 80,
     "score_high": 80,
     "score_source": "checklist",
     "usd_month": 0,
     "checked_on": "2026-10-11"
    }
   ],
   "price": {
    "free_plan": true,
    "usd_month": 0,
    "plans": "OpenCode is open source under MIT (s2). Optional Zen model access is pay as you go, with card fees passed on at 4.4% + $0.30 per top-up (s3).",
    "api": "Your chosen provider's rates, or Zen per-token prices, for example $0.50 / $3.00 per 1M input / output tokens for Gemini 3 Flash (s3).",
    "checked_on": "2026-10-11"
   },
   "how_we_rated": "From the SPEC checklist, using Artificial Analysis's Coding Agent Index v1.5 results and methodology, the arena.ai WebDev board, and OpenCode's website, GitHub page, release list and Zen docs, checked 11 Oct 2026. We did not run it ourselves.",
   "sources": [
    {
     "id": "s1",
     "title": "OpenCode home page",
     "url": "https://opencode.ai/?v=20261011",
     "checked": "2026-10-11"
    },
    {
     "id": "s2",
     "title": "GitHub: anomalyco/opencode",
     "url": "https://github.com/anomalyco/opencode?v=20261011",
     "checked": "2026-10-11"
    },
    {
     "id": "s3",
     "title": "OpenCode docs: Zen",
     "url": "https://opencode.ai/docs/zen/?v=20261011",
     "checked": "2026-10-11"
    },
    {
     "id": "s4",
     "title": "GitHub: OpenCode releases",
     "url": "https://github.com/anomalyco/opencode/releases?v=20261011",
     "checked": "2026-10-11"
    },
    {
     "id": "s5",
     "title": "Artificial Analysis: Claude Code vs OpenCode, Coding Agent Index v1.5",
     "url": "https://artificialanalysis.ai/agents/coding-agents/comparisons/claude-code-vs-opencode",
     "checked": "2026-10-11"
    },
    {
     "id": "s6",
     "title": "Artificial Analysis: Coding Agent Index v1.5 methodology",
     "url": "https://artificialanalysis.ai/methodology/coding-agents-benchmarking?v=20261011",
     "checked": "2026-10-11"
    },
    {
     "id": "s7",
     "title": "arena.ai Code Arena: WebDev leaderboard",
     "url": "https://arena.ai/leaderboard/code/webdev?v=20261011",
     "checked": "2026-10-11"
    }
   ],
   "checked": "2026-10-11"
  },
  {
   "slug": "openclaw",
   "url": "https://hayzlabs.com/reviews/openclaw/",
   "name": "OpenClaw",
   "maker": "OpenClaw Foundation",
   "kind": "open-source",
   "license": "open-source",
   "runs": "both",
   "released": "2026-10-10",
   "status": "first-look",
   "verdict": "Worth trying",
   "verdict_line": "Popular free personal agent you run on your own computer; powerful but risky, so set up sandboxing before you use it.",
   "grades": [],
   "price": {
    "free_plan": true,
    "usd_month": 0,
    "plans": "Free and open source under MIT; you pay for the model you plug in, or run a local model (s1).",
    "api": "Your chosen model provider's rates (s1).",
    "checked_on": "2026-10-11"
   },
   "how_we_rated": "First look from OpenClaw's GitHub page, release list and security docs, checked 11 Oct 2026. Not rated yet. We did not run it ourselves.",
   "sources": [
    {
     "id": "s1",
     "title": "GitHub: openclaw/openclaw",
     "url": "https://github.com/openclaw/openclaw?v=20261011",
     "checked": "2026-10-11"
    },
    {
     "id": "s2",
     "title": "OpenClaw releases on GitHub",
     "url": "https://github.com/openclaw/openclaw/releases?v=20261011",
     "checked": "2026-10-11"
    },
    {
     "id": "s3",
     "title": "OpenClaw docs: Gateway security",
     "url": "https://docs.openclaw.ai/gateway/security?v=20261011",
     "checked": "2026-10-11"
    }
   ],
   "checked": "2026-10-11"
  },
  {
   "slug": "n8n",
   "url": "https://hayzlabs.com/reviews/n8n/",
   "name": "n8n",
   "maker": "n8n",
   "kind": "app",
   "license": "source-available",
   "runs": "both",
   "released": "2026-10-09",
   "status": "first-look",
   "verdict": "Worth trying",
   "verdict_line": "Widely used tool for building AI automations and agents; free to self-host, cloud from 20 euros a month.",
   "grades": [],
   "price": {
    "free_plan": true,
    "usd_month": 0,
    "plans": "Community Edition free to self-host (s1). Cloud, billed yearly: Starter 20 euros/month for 2,500 executions; Pro 50 euros/month for 10,000; Business 667 euros/month (self-hosted); Enterprise on request (s1).",
    "api": "AI steps use your own model keys; the built-in Assistant uses monthly credits on cloud plans, or your own key when self-hosted (s1).",
    "checked_on": "2026-10-11"
   },
   "how_we_rated": "First look from n8n's pricing page, GitHub page, release list and self-hosting docs, checked 11 Oct 2026. Not rated yet. We did not run it ourselves.",
   "sources": [
    {
     "id": "s1",
     "title": "n8n pricing",
     "url": "https://n8n.io/pricing/?v=20261011",
     "checked": "2026-10-11"
    },
    {
     "id": "s2",
     "title": "GitHub: n8n-io/n8n",
     "url": "https://github.com/n8n-io/n8n?v=20261011",
     "checked": "2026-10-11"
    },
    {
     "id": "s3",
     "title": "n8n docs: isolate n8n (telemetry settings)",
     "url": "https://docs.n8n.io/deploy/host-n8n/configure-n8n/basic-configuration/configuration-examples/isolate-n8n",
     "checked": "2026-10-11"
    },
    {
     "id": "s4",
     "title": "n8n releases on GitHub",
     "url": "https://github.com/n8n-io/n8n/releases?v=20261011",
     "checked": "2026-10-11"
    }
   ],
   "checked": "2026-10-11"
  },
  {
   "slug": "openai-codex-cli",
   "url": "https://hayzlabs.com/reviews/openai-codex-cli/",
   "name": "OpenAI Codex CLI",
   "maker": "OpenAI",
   "kind": "open-source",
   "license": "open-source",
   "runs": "both",
   "released": "2026-10-09",
   "status": "graded",
   "verdict": "Worth trying",
   "verdict_line": "Close to the top coding-agent score at a fraction of the cost per task; needs ChatGPT Plus or an API key.",
   "grades": [
    {
     "category": "coding",
     "score": 70,
     "score_low": 70,
     "score_high": 70,
     "score_source": "checklist",
     "usd_month": 20,
     "checked_on": "2026-10-11"
    }
   ],
   "price": {
    "free_plan": false,
    "usd_month": 20,
    "plans": "Included with ChatGPT Plus $20/month, Pro from $100, Business $25/user monthly; the CLI also works with an API key at API rates (s5). Free and Go are listed only with the desktop app (s5).",
    "api": "API rates for the model your key can use (s5).",
    "checked_on": "2026-10-11"
   },
   "how_we_rated": "From the SPEC checklist, using Artificial Analysis's Coding Agent Index v1.5 results and methodology, Snorkel's Terminal-Bench 2.1 board, the arena.ai WebDev board and OpenAI's pricing, model and data pages, checked 11 Oct 2026. We did not run it ourselves.",
   "sources": [
    {
     "id": "s1",
     "title": "Artificial Analysis: Cursor Composer 2.5 on the Coding Agent Index (20 May 2026)",
     "url": "https://artificialanalysis.ai/articles/cursor-composer-2-5-coding-agent-index",
     "checked": "2026-10-11"
    },
    {
     "id": "s2",
     "title": "Artificial Analysis: Coding agents leaderboard",
     "url": "https://artificialanalysis.ai/agents/coding-agents?v=20261011",
     "checked": "2026-10-11"
    },
    {
     "id": "s3",
     "title": "Codex docs: Models",
     "url": "https://learn.chatgpt.com/docs/models?v=20261011",
     "checked": "2026-10-11"
    },
    {
     "id": "s4",
     "title": "arena.ai Code Arena: WebDev leaderboard",
     "url": "https://arena.ai/leaderboard/code/webdev?v=20261011",
     "checked": "2026-10-11"
    },
    {
     "id": "s5",
     "title": "Codex docs: Pricing",
     "url": "https://learn.chatgpt.com/docs/pricing?v=20261011",
     "checked": "2026-10-11"
    },
    {
     "id": "s6",
     "title": "OpenAI Help: How your data is used to improve model performance",
     "url": "https://help.openai.com/en/articles/5722486-how-your-data-is-used-to-improve-model-performance",
     "checked": "2026-10-11"
    },
    {
     "id": "s7",
     "title": "arena.ai Agent leaderboard",
     "url": "https://arena.ai/leaderboard/agent?v=20261011",
     "checked": "2026-10-11"
    },
    {
     "id": "s8",
     "title": "openai/codex on GitHub",
     "url": "https://github.com/openai/codex?v=20261011",
     "checked": "2026-10-11"
    },
    {
     "id": "s9",
     "title": "OpenAI: DevDay 2026 recap",
     "url": "https://openai.com/index/devday-2026-recap/?v=20261011",
     "checked": "2026-10-11"
    },
    {
     "id": "s10",
     "title": "openai/codex releases",
     "url": "https://github.com/openai/codex/releases?v=20261011",
     "checked": "2026-10-11"
    },
    {
     "id": "s11",
     "title": "Artificial Analysis: Claude Code vs Codex, Coding Agent Index v1.5",
     "url": "https://artificialanalysis.ai/agents/coding-agents/comparisons/claude-code-vs-codex",
     "checked": "2026-10-11"
    },
    {
     "id": "s12",
     "title": "Snorkel AI: Terminal-Bench 2.1 leaderboard (updated 1 Oct 2026)",
     "url": "https://snorkel.ai/leaderboard/terminal-bench-2-1/?v=20261011",
     "checked": "2026-10-11"
    },
    {
     "id": "s13",
     "title": "Artificial Analysis: Coding Agent Index v1.5 methodology",
     "url": "https://artificialanalysis.ai/methodology/coding-agents-benchmarking?v=20261011",
     "checked": "2026-10-11"
    }
   ],
   "checked": "2026-10-11"
  },
  {
   "slug": "cline",
   "url": "https://hayzlabs.com/reviews/cline/",
   "name": "Cline",
   "maker": "Cline Bot Inc.",
   "kind": "open-source",
   "license": "open-source",
   "runs": "both",
   "released": "2026-10-08",
   "status": "first-look",
   "verdict": "Worth trying",
   "verdict_line": "Free open-source coding agent; you pick the model and pay the provider. Quality depends on the model you choose.",
   "grades": [],
   "price": {
    "free_plan": true,
    "usd_month": 0,
    "plans": "Open Source free for individuals with no seat fee; you pay for model use, through Cline's provider or your own key. Enterprise custom (s2).",
    "api": "Your chosen provider's rates; Cline lists inference 'at cost or BYOK' (s2).",
    "checked_on": "2026-10-11"
   },
   "how_we_rated": "First look from Cline's GitHub page, pricing page and docs, checked 11 Oct 2026. Not rated yet. We did not run it ourselves.",
   "sources": [
    {
     "id": "s1",
     "title": "Cline on GitHub",
     "url": "https://github.com/cline/cline?v=20261011",
     "checked": "2026-10-11"
    },
    {
     "id": "s2",
     "title": "Cline pricing",
     "url": "https://cline.bot/pricing?v=20261011",
     "checked": "2026-10-11"
    },
    {
     "id": "s3",
     "title": "Cline docs: Telemetry",
     "url": "https://docs.cline.bot/enterprise-solutions/monitoring/telemetry",
     "checked": "2026-10-11"
    },
    {
     "id": "s4",
     "title": "Cline docs: Ollama and local models",
     "url": "https://docs.cline.bot/running-models-locally/overview",
     "checked": "2026-10-11"
    },
    {
     "id": "s5",
     "title": "Cline releases on GitHub",
     "url": "https://github.com/cline/cline/releases?v=20261011",
     "checked": "2026-10-11"
    }
   ],
   "checked": "2026-10-11"
  },
  {
   "slug": "comfyui",
   "url": "https://hayzlabs.com/reviews/comfyui/",
   "name": "ComfyUI",
   "maker": "Comfy Org",
   "kind": "open-source",
   "license": "open-source",
   "runs": "local",
   "released": "2026-10-08",
   "status": "first-look",
   "verdict": "Worth trying",
   "verdict_line": "Widely used free app for running open image and video models on your own computer; expect a learning curve.",
   "grades": [],
   "price": {
    "free_plan": true,
    "usd_month": 0,
    "plans": "Free and open source (GPL-3.0) to run on your own computer; Comfy Cloud is a paid hosted version, and optional partner API nodes for closed models are paid (s1).",
    "api": null,
    "checked_on": "2026-10-11"
   },
   "how_we_rated": "First look from the project's GitHub page and package listing, checked 11 Oct 2026. Not rated yet. We did not run it ourselves.",
   "sources": [
    {
     "id": "s1",
     "title": "ComfyUI on GitHub (Comfy-Org/ComfyUI)",
     "url": "https://github.com/Comfy-Org/ComfyUI?v=20261011",
     "checked": "2026-10-11"
    },
    {
     "id": "s2",
     "title": "PyPI: comfyui-frontend-package",
     "url": "https://pypi.org/project/comfyui-frontend-package/?v=20261011",
     "checked": "2026-10-11"
    }
   ],
   "checked": "2026-10-11"
  },
  {
   "slug": "hermes-agent",
   "url": "https://hayzlabs.com/reviews/hermes-agent/",
   "name": "Hermes Agent",
   "maker": "Nous Research",
   "kind": "open-source",
   "license": "open-source",
   "runs": "both",
   "released": "2026-10-08",
   "status": "first-look",
   "verdict": "Worth trying",
   "verdict_line": "Free personal agent that remembers your work and runs from the terminal or chat apps; sandbox it before real use.",
   "grades": [],
   "price": {
    "free_plan": true,
    "usd_month": 0,
    "plans": "Free and open source under MIT. Optional Nous Portal for hosted models: Free $0 (free models only), Plus $20, Super $100, Ultra $200 a month (s3).",
    "api": "Your chosen model provider's rates, or Nous Portal credits (s3).",
    "checked_on": "2026-10-11"
   },
   "how_we_rated": "First look from Nous Research's GitHub page, website, release list and security docs, and launch news, checked 11 Oct 2026. Not rated yet. We did not run it ourselves.",
   "sources": [
    {
     "id": "s1",
     "title": "GitHub: NousResearch/hermes-agent",
     "url": "https://github.com/NousResearch/hermes-agent?v=20261011",
     "checked": "2026-10-11"
    },
    {
     "id": "s2",
     "title": "The Next Web: Nous Research raises $90M to take its Hermes AI agent to businesses",
     "url": "https://thenextweb.com/news/nous-research-90m-1-5bn-hermes-agent-wsj?v=20261011",
     "checked": "2026-10-11"
    },
    {
     "id": "s3",
     "title": "Hermes Agent website and Nous Portal pricing",
     "url": "https://hermes-agent.nousresearch.com/?v=20261011",
     "checked": "2026-10-11"
    },
    {
     "id": "s4",
     "title": "Hermes Agent releases on GitHub",
     "url": "https://github.com/NousResearch/hermes-agent/releases?v=20261011",
     "checked": "2026-10-11"
    },
    {
     "id": "s5",
     "title": "Hermes Agent docs: Security",
     "url": "https://hermes-agent.nousresearch.com/docs/user-guide/security?v=20261011",
     "checked": "2026-10-11"
    }
   ],
   "checked": "2026-10-11"
  },
  {
   "slug": "ollama",
   "url": "https://hayzlabs.com/reviews/ollama/",
   "name": "Ollama",
   "maker": "Ollama",
   "kind": "open-source",
   "license": "open-source",
   "runs": "both",
   "released": "2026-10-08",
   "status": "graded",
   "verdict": "Worth trying",
   "verdict_line": "Free, private way to run open coding models like Qwen3.8 27B on your own computer; skip it if you need frontier quality.",
   "grades": [
    {
     "category": "coding",
     "score": 80,
     "score_low": 80,
     "score_high": 80,
     "score_source": "checklist",
     "usd_month": 0,
     "checked_on": "2026-10-18"
    }
   ],
   "price": {
    "free_plan": true,
    "usd_month": 0,
    "plans": "Local use free and unlimited; optional cloud models: Pro $20/month ($200/year), Max $100/month, Team $500/month (s4)",
    "api": null,
    "checked_on": "2026-10-11"
   },
   "how_we_rated": "From the SPEC checklist, using an independent speed measurement, arena.ai ranks of the models it can run, and Ollama's own pricing, FAQ and release pages, checked 11 Oct 2026. We did not run it ourselves.",
   "sources": [
    {
     "id": "s1",
     "title": "TerminalBytes: Ollama vs llama.cpp vs MLX on Mac (2026), 20 Sep 2026",
     "url": "https://terminalbytes.com/ollama-vs-llama-cpp-vs-mlx-mac-2026/",
     "checked": "2026-10-18"
    },
    {
     "id": "s2",
     "title": "Ollama model library (newest) and qwen3.8 model page",
     "url": "https://ollama.com/library/qwen3.8",
     "checked": "2026-10-11"
    },
    {
     "id": "s3",
     "title": "arena.ai Code Arena: WebDev leaderboard",
     "url": "https://arena.ai/leaderboard/code/webdev?v=20261011",
     "checked": "2026-10-11"
    },
    {
     "id": "s4",
     "title": "Ollama pricing",
     "url": "https://ollama.com/pricing?v=20261011",
     "checked": "2026-10-11"
    },
    {
     "id": "s5",
     "title": "Ollama FAQ",
     "url": "https://docs.ollama.com/faq",
     "checked": "2026-10-11"
    },
    {
     "id": "s6",
     "title": "Ollama releases on GitHub",
     "url": "https://github.com/ollama/ollama/releases?v=20261011",
     "checked": "2026-10-11"
    },
    {
     "id": "s7",
     "title": "Ollama repository on GitHub",
     "url": "https://github.com/ollama/ollama",
     "checked": "2026-10-11"
    },
    {
     "id": "s8",
     "title": "arena.ai Text: Coding leaderboard",
     "url": "https://arena.ai/leaderboard/chat/text/coding?v=20261011",
     "checked": "2026-10-11"
    }
   ],
   "checked": "2026-10-11"
  },
  {
   "slug": "chatgpt-gpt-6",
   "url": "https://hayzlabs.com/reviews/chatgpt-gpt-6/",
   "name": "ChatGPT (GPT-6)",
   "maker": "OpenAI",
   "kind": "app",
   "license": "proprietary",
   "runs": "cloud",
   "released": "2026-10-07",
   "status": "graded",
   "verdict": "Depends",
   "verdict_line": "Good for agent work and coding on Plus with GPT-6 Astra. For writing, Gemini AI Pro and Claude Pro score higher.",
   "grades": [
    {
     "category": "agents",
     "score": 70,
     "score_low": 70,
     "score_high": 70,
     "score_source": "checklist",
     "usd_month": 20,
     "checked_on": "2026-10-11"
    },
    {
     "category": "coding",
     "score": 70,
     "score_low": 70,
     "score_high": 70,
     "score_source": "checklist",
     "usd_month": 20,
     "checked_on": "2026-10-11"
    },
    {
     "category": "research",
     "score": 60,
     "score_low": 60,
     "score_high": 60,
     "score_source": "checklist",
     "usd_month": 20,
     "checked_on": "2026-10-11"
    },
    {
     "category": "writing",
     "score": 50,
     "score_low": 50,
     "score_high": 50,
     "score_source": "checklist",
     "usd_month": 20,
     "checked_on": "2026-10-11"
    }
   ],
   "price": {
    "free_plan": true,
    "usd_month": 20,
    "plans": "Free; Go (may include ads; no price shown on the pricing page); Plus $20/month; Pro $100, $200 or $500/month (s6, s7, s13)",
    "api": null,
    "checked_on": "2026-10-11"
   },
   "how_we_rated": "From the SPEC checklist, scored per plan with one rule for every app: each plan is scored on the best model it lets you choose, as documented on the maker's own plan and help pages, and we name it; a default the maker names does not lower the score. Does the job is 2 when that model sits in the top band of Arena's ranks for the job, else 1; underlying quality is 2 for the top band, 1 for the next band, 0 below. The score comes from the higher-scoring plan; a tie goes to the cheaper one. Ranks are from arena.ai, and plans and privacy from OpenAI's own pages, checked 11 Oct 2026. We did not run it ourselves. Where we point to Claude, we also checked our reviews for the best non-Anthropic option: Gemini for writing and LM Studio for coding; for hard questions, none scores above ChatGPT. Chat jobs (writing, hard questions) use the models in chat; agents and coding also count ChatGPT Work, the agent mode inside the app, which OpenAI's pricing page lists for Plus on desktop, web and mobile (s6) and which Plus runs on GPT-6 Astra (s15); Codex has its own page.",
   "sources": [
    {
     "id": "s1",
     "title": "arena.ai Code Arena: WebDev leaderboard",
     "url": "https://arena.ai/leaderboard/code/webdev?v=20261011",
     "checked": "2026-10-11"
    },
    {
     "id": "s2",
     "title": "arena.ai Agent leaderboard",
     "url": "https://arena.ai/leaderboard/agent?v=20261011",
     "checked": "2026-10-11"
    },
    {
     "id": "s3",
     "title": "arena.ai Text: Expert leaderboard",
     "url": "https://arena.ai/leaderboard/chat/text/expert?v=20261011",
     "checked": "2026-10-11"
    },
    {
     "id": "s4",
     "title": "arena.ai Text: Creative Writing leaderboard",
     "url": "https://arena.ai/leaderboard/chat/text/creative-writing?v=20261011",
     "checked": "2026-10-11"
    },
    {
     "id": "s5",
     "title": "arena.ai Text: Coding leaderboard",
     "url": "https://arena.ai/leaderboard/chat/text/coding?v=20261011",
     "checked": "2026-10-11"
    },
    {
     "id": "s6",
     "title": "ChatGPT pricing and plan comparison",
     "url": "https://chatgpt.com/pricing?v=20261011",
     "checked": "2026-10-11"
    },
    {
     "id": "s7",
     "title": "OpenAI Help: What is ChatGPT Plus?",
     "url": "https://help.openai.com/en/articles/6950777-what-is-chatgpt-plus",
     "checked": "2026-10-11"
    },
    {
     "id": "s8",
     "title": "OpenAI Help: ChatGPT release notes",
     "url": "https://help.openai.com/en/articles/6825453-chatgpt-release-notes",
     "checked": "2026-10-11"
    },
    {
     "id": "s9",
     "title": "OpenAI Help: How your data is used to improve model performance",
     "url": "https://help.openai.com/en/articles/5722486-how-your-data-is-used-to-improve-model-performance",
     "checked": "2026-10-11"
    },
    {
     "id": "s10",
     "title": "OpenAI Help: Data controls FAQ",
     "url": "https://help.openai.com/en/articles/7730893-data-controls-faq",
     "checked": "2026-10-11"
    },
    {
     "id": "s11",
     "title": "Tom's Guide: ChatGPT-6 vs Claude Opus 5.5 with 5 everyday prompts (24 Sep 2026)",
     "url": "https://www.tomsguide.com/ai/i-tested-chatgpt-6-vs-claude-opus-5-5-with-5-everyday-prompts-it-wasnt-even-close",
     "checked": "2026-10-11"
    },
    {
     "id": "s12",
     "title": "OpenAI Help: Using Codex with your ChatGPT plan",
     "url": "https://help.openai.com/en/articles/11369540-using-codex-with-your-chatgpt-plan",
     "checked": "2026-10-11"
    },
    {
     "id": "s13",
     "title": "OpenAI Help: About ChatGPT Pro tiers",
     "url": "https://help.openai.com/en/articles/9793128-about-chatgpt-pro-tiers",
     "checked": "2026-10-11"
    },
    {
     "id": "s14",
     "title": "arena.ai Agent Alignment Index",
     "url": "https://arena.ai/leaderboard/agent/alignment?v=20261011",
     "checked": "2026-10-11"
    },
    {
     "id": "s15",
     "title": "OpenAI Help: GPT-6 and other models in ChatGPT",
     "url": "https://help.openai.com/en/articles/11909943-gpt-56-and-gpt-6-pro-in-chatgpt",
     "checked": "2026-10-11"
    }
   ],
   "checked": "2026-10-11"
  },
  {
   "slug": "claude-haiku-5-5",
   "url": "https://hayzlabs.com/reviews/claude-haiku-5-5/",
   "name": "Claude Haiku 5.5",
   "maker": "Anthropic",
   "kind": "model",
   "license": "proprietary",
   "runs": "cloud",
   "released": "2026-10-07",
   "status": "graded",
   "verdict": "Worth trying",
   "verdict_line": "Very cheap small model for summaries, sorting and coding subagents; use Sonnet 5.5 for hard coding work.",
   "grades": [
    {
     "category": "coding",
     "score": 44,
     "score_low": 42,
     "score_high": 46,
     "score_source": "model",
     "score_inputs": [
      "Arena"
     ],
     "usd_month": null,
     "checked_on": "2026-10-11"
    }
   ],
   "price": {
    "free_plan": false,
    "usd_month": null,
    "plans": "Anthropic's launch page names API and cloud availability; it does not say which Claude app plans include it (s3).",
    "api": "$0.10 / $0.50 per 1M input / output tokens for prompts up to 100K tokens; $0.50 / $2.50 above that (s3)",
    "checked_on": "2026-10-11"
   },
   "how_we_rated": "From the arena.ai WebDev leaderboard, with the Artificial Analysis Intelligence Index for context, checked 11 Oct 2026. We did not run it ourselves.",
   "sources": [
    {
     "id": "s1",
     "title": "arena.ai Code Arena: WebDev leaderboard",
     "url": "https://arena.ai/leaderboard/code/webdev?v=20261011",
     "checked": "2026-10-11"
    },
    {
     "id": "s2",
     "title": "arena.ai Text: Coding leaderboard",
     "url": "https://arena.ai/leaderboard/chat/text/coding?v=20261011",
     "checked": "2026-10-11"
    },
    {
     "id": "s3",
     "title": "Anthropic: Introducing Claude Haiku 5.5",
     "url": "https://www.anthropic.com/claude-haiku-5-5",
     "checked": "2026-10-11"
    },
    {
     "id": "s4",
     "title": "Artificial Analysis LLM leaderboard",
     "url": "https://artificialanalysis.ai/leaderboards/models?v=20261011",
     "checked": "2026-10-11"
    },
    {
     "id": "s5",
     "title": "Claude privacy: is my data used for model training (commercial)",
     "url": "https://privacy.claude.com/en/articles/7996868-is-my-data-used-for-model-training",
     "checked": "2026-10-11"
    },
    {
     "id": "s6",
     "title": "arena.ai Agent leaderboard",
     "url": "https://arena.ai/leaderboard/agent?v=20261011",
     "checked": "2026-10-11"
    },
    {
     "id": "s7",
     "title": "Deccan Chronicle (Reuters): Anthropic launches third Claude 5.5 model",
     "url": "https://www.deccanchronicle.com/technology/anthropic-launches-third-claude-55-model-expanding-ai-lineup-before-planned-ipo-1994077",
     "checked": "2026-10-11"
    },
    {
     "id": "s8",
     "title": "YourStory: Claude Haiku 5.5 debuts as Anthropic's faster small AI model",
     "url": "https://yourstory.com/ai-story/claude-haiku-5-5-anthropic-small-ai-model",
     "checked": "2026-10-11"
    }
   ],
   "checked": "2026-10-11"
  },
  {
   "slug": "cursor",
   "url": "https://hayzlabs.com/reviews/cursor/",
   "name": "Cursor",
   "maker": "Cursor (Anysphere, part of SpaceX)",
   "kind": "dev-tool",
   "license": "proprietary",
   "runs": "both",
   "released": "2026-10-06",
   "status": "first-look",
   "verdict": "Worth trying",
   "verdict_line": "AI code editor with a cheap in-house model; default model unstated and trains on code unless Privacy Mode is on.",
   "grades": [],
   "price": {
    "free_plan": true,
    "usd_month": 8,
    "plans": "Free; Plus $8/month; Premium $30; Super $100; Ultra $200; Teams from $30/user. Claude, GPT and Gemini usage is billed separately at API list price (s7).",
    "api": "Other Models pool billed at each model's API rate; Composer 2.5 $0.50 / $2.50 per 1M input / output tokens (s5).",
    "checked_on": "2026-10-19"
   },
   "how_we_rated": "First look from Cursor's pricing, model and data pages and the arena.ai WebDev and Agent boards, with the May 2026 Artificial Analysis coding agent article as dated context, checked 11 Oct 2026. Not rated yet. We did not run it ourselves.",
   "sources": [
    {
     "id": "s1",
     "title": "Artificial Analysis: Cursor Composer 2.5 on the Coding Agent Index (20 May 2026)",
     "url": "https://artificialanalysis.ai/articles/cursor-composer-2-5-coding-agent-index",
     "checked": "2026-10-11"
    },
    {
     "id": "s2",
     "title": "Artificial Analysis: Coding agents leaderboard",
     "url": "https://artificialanalysis.ai/agents/coding-agents?v=20261011",
     "checked": "2026-10-11"
    },
    {
     "id": "s3",
     "title": "Cursor docs: Models",
     "url": "https://cursor.com/docs/models?v=20261011",
     "checked": "2026-10-11"
    },
    {
     "id": "s4",
     "title": "arena.ai Code Arena: WebDev leaderboard",
     "url": "https://arena.ai/leaderboard/code/webdev?v=20261011",
     "checked": "2026-10-11"
    },
    {
     "id": "s5",
     "title": "Cursor docs: Pricing",
     "url": "https://cursor.com/docs/account/pricing?v=20261011",
     "checked": "2026-10-11"
    },
    {
     "id": "s6",
     "title": "Cursor: Data use",
     "url": "https://cursor.com/data-use?v=20261011",
     "checked": "2026-10-11"
    },
    {
     "id": "s7",
     "title": "Cursor pricing",
     "url": "https://cursor.com/pricing?v=1019",
     "checked": "2026-10-19"
    },
    {
     "id": "s8",
     "title": "arena.ai Agent leaderboard",
     "url": "https://arena.ai/leaderboard/agent?v=20261011",
     "checked": "2026-10-11"
    },
    {
     "id": "s9",
     "title": "Cursor: Security",
     "url": "https://cursor.com/security?v=20261011",
     "checked": "2026-10-11"
    },
    {
     "id": "s10",
     "title": "Cursor changelog",
     "url": "https://cursor.com/changelog?v=20261011",
     "checked": "2026-10-11"
    },
    {
     "id": "s11",
     "title": "Cursor blog: Cursor is now a part of SpaceX (14 Aug 2026)",
     "url": "https://cursor.com/blog/joining-spacex?v=20261011",
     "checked": "2026-10-11"
    },
    {
     "id": "s12",
     "title": "OpenAI Help Center: Using OpenAI models in Cursor",
     "url": "https://help.openai.com/en/articles/20001506-using-openai-models-in-cursor",
     "checked": "2026-10-19"
    }
   ],
   "checked": "2026-10-11"
  },
  {
   "slug": "embeddinggemma-2",
   "url": "https://hayzlabs.com/reviews/embeddinggemma-2/",
   "name": "EmbeddingGemma 2",
   "maker": "Google",
   "kind": "open-source",
   "license": "open-weights",
   "runs": "local",
   "released": "2026-10-06",
   "status": "first-look",
   "verdict": "Wait",
   "verdict_line": "Free, tiny Google search model for text, images, audio and video; scores are Google's own, so wait for outside results.",
   "grades": [],
   "price": {
    "free_plan": true,
    "usd_month": 0,
    "plans": "Free to download and run under Apache 2.0 (s1).",
    "api": null,
    "checked_on": "2026-10-11"
   },
   "how_we_rated": "First look from Google's model card and launch coverage, checked 11 Oct 2026. Not rated yet. We did not run it ourselves.",
   "sources": [
    {
     "id": "s1",
     "title": "Hugging Face: google/embeddinggemma-2 model card",
     "url": "https://huggingface.co/google/embeddinggemma-2?v=20261011",
     "checked": "2026-10-11"
    },
    {
     "id": "s2",
     "title": "SiliconANGLE: Google expands EmbeddingGemma beyond text to images, audio and video",
     "url": "https://siliconangle.com/2026/10/06/google-expands-embeddinggemma-beyond-text-to-images-audio-and-video/",
     "checked": "2026-10-11"
    }
   ],
   "checked": "2026-10-11"
  },
  {
   "slug": "mistral-large-4",
   "url": "https://hayzlabs.com/reviews/mistral-large-4/",
   "name": "Mistral Large 4",
   "maker": "Mistral AI",
   "kind": "model",
   "license": "open-weights",
   "runs": "cloud",
   "released": "2026-10-06",
   "status": "graded",
   "verdict": "Wait",
   "verdict_line": "Mistral's 1T preview scores no higher than 72 in any job; weights due 31 Oct. Use GLM-5.3 or Kimi K3 now.",
   "grades": [
    {
     "category": "coding",
     "score": 36,
     "score_low": 34,
     "score_high": 38,
     "score_source": "model",
     "score_inputs": [
      "Arena"
     ],
     "usd_month": null,
     "checked_on": "2026-10-11"
    },
    {
     "category": "agents",
     "score": null,
     "score_low": null,
     "score_high": null,
     "score_source": "model",
     "score_inputs": [
      "Arena"
     ],
     "arena_rank": 39,
     "usd_month": null,
     "checked_on": "2026-10-11"
    },
    {
     "category": "research",
     "score": 62,
     "score_low": 56,
     "score_high": 67,
     "score_source": "model",
     "score_inputs": [
      "Arena",
      "Epoch AI"
     ],
     "usd_month": null,
     "checked_on": "2026-10-11"
    },
    {
     "category": "writing",
     "score": 59,
     "score_low": 55,
     "score_high": 63,
     "score_source": "model",
     "score_inputs": [
      "Arena"
     ],
     "usd_month": null,
     "checked_on": "2026-10-11"
    }
   ],
   "price": {
    "free_plan": false,
    "usd_month": null,
    "plans": "API only during the preview, through Mistral Studio (s1).",
    "api": "List $1.36 / $4.18 per 1M input / output tokens, on sale at $0.68 / $2.09; cached input $0.14 list, $0.07 on sale (s1)",
    "checked_on": "2026-10-11"
   },
   "how_we_rated": "From arena.ai leaderboards and the Artificial Analysis Intelligence Index, checked 11 Oct 2026, with prices from Mistral's model page. It is a preview still in training, so ratings may move soon. We did not run it ourselves.",
   "sources": [
    {
     "id": "s1",
     "title": "Mistral docs: Mistral Large 4 model card",
     "url": "https://docs.mistral.ai/models/mistral-large-4-0",
     "checked": "2026-10-11"
    },
    {
     "id": "s2",
     "title": "The Next Web: Europe's Mistral launches Large 4",
     "url": "https://thenextweb.com/news/mistral-releases-large-4-a-1-trillion-parameter-open-weight-ai-model",
     "checked": "2026-10-11"
    },
    {
     "id": "s3",
     "title": "BetaNews: Mistral opens trillion-parameter Large 4 preview",
     "url": "https://betanews.com/article/mistral-large-4-preview-open-weights/",
     "checked": "2026-10-11"
    },
    {
     "id": "s4",
     "title": "Mistral Help Center: opting out of data use for training",
     "url": "https://help.mistral.ai/en/articles/455207-can-i-opt-out-of-my-input-or-output-data-being-used-for-training",
     "checked": "2026-10-11"
    },
    {
     "id": "s5",
     "title": "arena.ai Text: Coding leaderboard",
     "url": "https://arena.ai/leaderboard/chat/text/coding?v=20261011",
     "checked": "2026-10-11"
    },
    {
     "id": "s6",
     "title": "arena.ai Agent leaderboard",
     "url": "https://arena.ai/leaderboard/agent?v=20261011",
     "checked": "2026-10-11"
    },
    {
     "id": "s7",
     "title": "Mistral changelog",
     "url": "https://docs.mistral.ai/resources/changelogs",
     "checked": "2026-10-11"
    },
    {
     "id": "s8",
     "title": "Hugging Face: mistralai/Mistral-Large-4-1T-A52B",
     "url": "https://huggingface.co/mistralai/Mistral-Large-4-1T-A52B?v=20261011",
     "checked": "2026-10-11"
    },
    {
     "id": "s9",
     "title": "arena.ai Code Arena: WebDev leaderboard",
     "url": "https://arena.ai/leaderboard/code/webdev?v=20261011",
     "checked": "2026-10-11"
    },
    {
     "id": "s10",
     "title": "arena.ai Text: Expert leaderboard",
     "url": "https://arena.ai/leaderboard/chat/text/expert?v=20261011",
     "checked": "2026-10-11"
    },
    {
     "id": "s11",
     "title": "Artificial Analysis LLM leaderboard",
     "url": "https://artificialanalysis.ai/leaderboards/models?v=20261011",
     "checked": "2026-10-11"
    },
    {
     "id": "s12",
     "title": "arena.ai Text: Creative Writing leaderboard",
     "url": "https://arena.ai/leaderboard/chat/text/creative-writing?v=20261011",
     "checked": "2026-10-11"
    },
    {
     "id": "s13",
     "title": "arena.ai Text: Instruction Following leaderboard",
     "url": "https://arena.ai/leaderboard/chat/text/instruction-following?v=20261011",
     "checked": "2026-10-11"
    }
   ],
   "checked": "2026-10-11"
  },
  {
   "slug": "nano-banana-2-1",
   "url": "https://hayzlabs.com/reviews/nano-banana-2-1/",
   "name": "Nano Banana 2.1",
   "maker": "Google",
   "kind": "model",
   "license": "proprietary",
   "runs": "cloud",
   "released": "2026-10-06",
   "status": "graded",
   "verdict": "Use it",
   "verdict_line": "Near-top image quality for about 3 cents an image, free to try in Gemini. Skip only if you need the very best edits.",
   "grades": [
    {
     "category": "images",
     "score": 73,
     "score_low": 71,
     "score_high": 75,
     "score_source": "model",
     "score_inputs": [
      "Arena"
     ],
     "usd_month": 0,
     "checked_on": "2026-10-11"
    }
   ],
   "price": {
    "free_plan": true,
    "usd_month": 0,
    "plans": "Gemini app free plan includes image generation and editing (Google does not name the model version per plan); Google AI Plus $4.99/month for 2x usage (s6)",
    "api": "$1.50 per 1M input tokens; images $0.0336 at 1K, $0.0504 at 2K, $0.113 at 4K; Batch half price (s5)",
    "checked_on": "2026-10-11"
   },
   "how_we_rated": "From the arena.ai and Artificial Analysis image leaderboards, checked 11 Oct 2026. We did not run it ourselves.",
   "sources": [
    {
     "id": "s1",
     "title": "arena.ai Text to Image leaderboard",
     "url": "https://arena.ai/leaderboard/image/text-to-image?v=20261011",
     "checked": "2026-10-11"
    },
    {
     "id": "s2",
     "title": "arena.ai Image Edit leaderboard (Single Image Edit)",
     "url": "https://arena.ai/leaderboard/image/image-edit",
     "checked": "2026-10-11"
    },
    {
     "id": "s3",
     "title": "Artificial Analysis text-to-image leaderboard",
     "url": "https://artificialanalysis.ai/image/leaderboard/text-to-image?v=20261011",
     "checked": "2026-10-11"
    },
    {
     "id": "s4",
     "title": "Artificial Analysis image editing leaderboard",
     "url": "https://artificialanalysis.ai/image/leaderboard/editing?v=20261011",
     "checked": "2026-10-11"
    },
    {
     "id": "s5",
     "title": "Gemini API pricing",
     "url": "https://ai.google.dev/gemini-api/docs/pricing?v=20261011",
     "checked": "2026-10-11"
    },
    {
     "id": "s6",
     "title": "Google AI plans (Gemini subscriptions)",
     "url": "https://gemini.google/subscriptions/",
     "checked": "2026-10-11"
    },
    {
     "id": "s7",
     "title": "Gemini Apps Privacy Hub",
     "url": "https://support.google.com/gemini/answer/13594961?hl=en",
     "checked": "2026-10-11"
    },
    {
     "id": "s8",
     "title": "Google DeepMind: Nano Banana (Gemini Image Flash)",
     "url": "https://deepmind.google/models/gemini-image/flash/",
     "checked": "2026-10-11"
    },
    {
     "id": "s9",
     "title": "Gemini API changelog",
     "url": "https://ai.google.dev/gemini-api/docs/changelog?v=20261011",
     "checked": "2026-10-11"
    }
   ],
   "checked": "2026-10-11"
  },
  {
   "slug": "flux-3-image",
   "url": "https://hayzlabs.com/reviews/flux-3-image/",
   "name": "FLUX 3 Image",
   "maker": "Black Forest Labs",
   "kind": "model",
   "license": "proprietary",
   "runs": "cloud",
   "released": "2026-10-01",
   "status": "graded",
   "verdict": "Worth trying",
   "verdict_line": "Top 10 image model that takes up to ten reference images, about 5 cents each. Plain prompts: Nano Banana 2.1 is cheaper.",
   "grades": [
    {
     "category": "images",
     "score": null,
     "score_low": null,
     "score_high": null,
     "score_source": null,
     "usd_month": null,
     "checked_on": "2026-10-11"
    }
   ],
   "price": {
    "free_plan": false,
    "usd_month": null,
    "plans": "Pay as you go through the BFL API; no subscription needed (s4).",
    "api": "$0.041 per 768px image, $0.048 at 1K, $0.100 at 2K, $0.607 at 4K (s4)",
    "checked_on": "2026-10-11"
   },
   "how_we_rated": "From the Artificial Analysis text-to-image leaderboard, checked 11 Oct 2026; arena.ai does not list it yet. We did not run it ourselves.",
   "sources": [
    {
     "id": "s1",
     "title": "Artificial Analysis text-to-image leaderboard",
     "url": "https://artificialanalysis.ai/image/leaderboard/text-to-image?v=20261011",
     "checked": "2026-10-11"
    },
    {
     "id": "s2",
     "title": "arena.ai Text to Image leaderboard",
     "url": "https://arena.ai/leaderboard/image/text-to-image?v=20261011",
     "checked": "2026-10-11"
    },
    {
     "id": "s3",
     "title": "Artificial Analysis image editing leaderboard",
     "url": "https://artificialanalysis.ai/image/leaderboard/editing?v=20261011",
     "checked": "2026-10-11"
    },
    {
     "id": "s4",
     "title": "BFL API pricing",
     "url": "https://docs.bfl.ai/quick_start/pricing?v=20261011",
     "checked": "2026-10-11"
    },
    {
     "id": "s5",
     "title": "BFL release notes",
     "url": "https://docs.bfl.ml/release-notes.md",
     "checked": "2026-10-11"
    },
    {
     "id": "s6",
     "title": "Black Forest Labs privacy policy",
     "url": "https://bfl.ai/legal/privacy-policy",
     "checked": "2026-10-11"
    }
   ],
   "checked": "2026-10-11"
  },
  {
   "slug": "cohere-embed-5-pro",
   "url": "https://hayzlabs.com/reviews/cohere-embed-5-pro/",
   "name": "Cohere Embed 5 Pro",
   "maker": "Cohere",
   "kind": "model",
   "license": "proprietary",
   "runs": "cloud",
   "released": "2026-09-30",
   "status": "first-look",
   "verdict": "Wait",
   "verdict_line": "Cohere's new search embedding model for text, images and PDFs. No independent scores yet; wait for MTEB results.",
   "grades": [],
   "price": {
    "free_plan": true,
    "usd_month": null,
    "plans": "Trial keys are free but rate limited and not for production or commercial use (s2). Dedicated Model Vault instances from $3.00 an hour or $2,000 a month (s2). Cohere's pricing page lists no per-token rate for it.",
    "api": null,
    "checked_on": "2026-10-11"
   },
   "how_we_rated": "First look from Cohere's pages and launch coverage, checked 11 Oct 2026. Not rated yet. We did not run it ourselves.",
   "sources": [
    {
     "id": "s1",
     "title": "Cohere docs: models overview",
     "url": "https://docs.cohere.com/docs/models?v=20261011",
     "checked": "2026-10-11"
    },
    {
     "id": "s2",
     "title": "Cohere pricing",
     "url": "https://cohere.com/pricing?v=20261011",
     "checked": "2026-10-11"
    },
    {
     "id": "s3",
     "title": "Cohere data usage policy",
     "url": "https://cohere.com/enterprise-data-commitments",
     "checked": "2026-10-11"
    },
    {
     "id": "s4",
     "title": "The New Stack: Cohere Embed 5 Pro and Fast",
     "url": "https://thenewstack.io/cohere-embed-pro-fast/",
     "checked": "2026-10-11"
    },
    {
     "id": "s5",
     "title": "AI Magazine: Cohere Embed 5",
     "url": "https://aimagazine.com/news/cohere-embed-5-what-are-frontier-embedding-models",
     "checked": "2026-10-11"
    }
   ],
   "checked": "2026-10-11"
  },
  {
   "slug": "gemini-4-argon",
   "url": "https://hayzlabs.com/reviews/gemini-4-argon/",
   "name": "Gemini 4 Argon",
   "maker": "Google",
   "kind": "model",
   "license": "proprietary",
   "runs": "cloud",
   "released": "2026-09-30",
   "status": "graded",
   "verdict": "Use it",
   "verdict_line": "Google's top model, first in Arena's text votes, now in Google AI Plus from $4.99/month; votes are still early.",
   "grades": [
    {
     "category": "writing",
     "score": 99,
     "score_low": 93,
     "score_high": 100,
     "score_source": "model",
     "score_inputs": [
      "Arena"
     ],
     "usd_month": 4.99,
     "checked_on": "2026-10-20"
    },
    {
     "category": "research",
     "score": 95,
     "score_low": 88,
     "score_high": 98,
     "score_source": "model",
     "score_inputs": [
      "Arena",
      "Epoch AI"
     ],
     "usd_month": 4.99,
     "checked_on": "2026-10-20"
    },
    {
     "category": "coding",
     "score": 64,
     "score_low": 61,
     "score_high": 68,
     "score_source": "model",
     "score_inputs": [
      "Arena"
     ],
     "usd_month": 4.99,
     "checked_on": "2026-10-20"
    },
    {
     "category": "agents",
     "score": null,
     "score_low": null,
     "score_high": null,
     "score_source": "model",
     "score_inputs": [
      "Arena"
     ],
     "arena_rank": 6,
     "usd_month": 4.99,
     "checked_on": "2026-10-20"
    }
   ],
   "price": {
    "free_plan": false,
    "usd_month": 4.99,
    "plans": "Google AI Plus $4.99/month, Google AI Pro $19.99/month and Google AI Ultra from $99.99/month all include it; the Free plan does not (s2).",
    "api": "$2 / $10 per 1M input / output tokens until 30 Nov 2026, then $4 / $20 from 1 Dec 2026 (s3)",
    "checked_on": "2026-10-20"
   },
   "how_we_rated": "From Arena's open leaderboard dataset (Text published 15 Oct, WebDev and Agent published 19 Oct) and the Artificial Analysis Intelligence Index, plus Google's plan and price pages, checked 20 Oct 2026. Its votes are still early, so its ranks may move. We did not run it ourselves.",
   "sources": [
    {
     "id": "s1",
     "title": "Google: Gemini 4 Argon",
     "url": "https://blog.google/innovation-and-ai/models-and-research/gemini-models/gemini-4-argon/",
     "checked": "2026-10-11"
    },
    {
     "id": "s2",
     "title": "Google AI plans (Gemini subscriptions)",
     "url": "https://gemini.google/subscriptions/",
     "checked": "2026-10-20"
    },
    {
     "id": "s3",
     "title": "Gemini API pricing",
     "url": "https://ai.google.dev/gemini-api/docs/pricing",
     "checked": "2026-10-18"
    },
    {
     "id": "s4",
     "title": "Artificial Analysis LLM leaderboard",
     "url": "https://artificialanalysis.ai/leaderboards/models?v=20261011",
     "checked": "2026-10-11"
    },
    {
     "id": "s5",
     "title": "arena.ai Agent leaderboard",
     "url": "https://arena.ai/leaderboard/agent?v=20261011",
     "checked": "2026-10-11"
    },
    {
     "id": "s6",
     "title": "arena.ai Text leaderboard (Overall)",
     "url": "https://arena.ai/leaderboard/chat/text?v=20261011",
     "checked": "2026-10-11"
    },
    {
     "id": "s7",
     "title": "arena.ai Text: Creative Writing leaderboard",
     "url": "https://arena.ai/leaderboard/chat/text/creative-writing?v=20261011",
     "checked": "2026-10-11"
    },
    {
     "id": "s8",
     "title": "arena.ai Text: Instruction Following leaderboard",
     "url": "https://arena.ai/leaderboard/chat/text/instruction-following?v=20261011",
     "checked": "2026-10-11"
    },
    {
     "id": "s9",
     "title": "arena.ai Text: Coding leaderboard",
     "url": "https://arena.ai/leaderboard/chat/text/coding?v=20261011",
     "checked": "2026-10-11"
    },
    {
     "id": "s10",
     "title": "arena.ai Agent Alignment Index",
     "url": "https://arena.ai/leaderboard/agent/alignment?v=20261011",
     "checked": "2026-10-11"
    },
    {
     "id": "s11",
     "title": "Gemini API models",
     "url": "https://ai.google.dev/gemini-api/docs/models",
     "checked": "2026-10-11"
    },
    {
     "id": "s12",
     "title": "arena.ai Code Arena: WebDev leaderboard",
     "url": "https://arena.ai/leaderboard/code/webdev?v=20261011",
     "checked": "2026-10-11"
    },
    {
     "id": "s13",
     "title": "Arena open leaderboard dataset: text_style_control, latest split (published 2026-10-15), CC BY 4.0",
     "url": "https://huggingface.co/datasets/lmarena-ai/leaderboard-dataset/viewer/text_style_control/latest",
     "checked": "2026-10-20"
    },
    {
     "id": "s14",
     "title": "Arena open leaderboard dataset: webdev, latest split (published 2026-10-19), CC BY 4.0",
     "url": "https://huggingface.co/datasets/lmarena-ai/leaderboard-dataset/viewer/webdev/latest",
     "checked": "2026-10-20"
    },
    {
     "id": "s15",
     "title": "Arena open leaderboard dataset: agent, latest split (published 2026-10-19), CC BY 4.0",
     "url": "https://huggingface.co/datasets/lmarena-ai/leaderboard-dataset/viewer/agent/latest",
     "checked": "2026-10-20"
    }
   ],
   "checked": "2026-10-20"
  },
  {
   "slug": "gpt-6-1-sol",
   "url": "https://hayzlabs.com/reviews/gpt-6-1-sol/",
   "name": "GPT-6.1 Sol",
   "maker": "OpenAI",
   "kind": "model",
   "license": "proprietary",
   "runs": "cloud",
   "released": "2026-09-29",
   "status": "graded",
   "verdict": "Use it",
   "verdict_line": "Near-top coding and agent work at a fifth of GPT-6 Astra's price; pick Claude Opus 5.5 or Gemini for creative writing.",
   "grades": [
    {
     "category": "agents",
     "score": null,
     "score_low": null,
     "score_high": null,
     "score_source": "model",
     "score_inputs": [
      "Arena"
     ],
     "arena_rank": 4,
     "usd_month": 20,
     "checked_on": "2026-10-11"
    },
    {
     "category": "coding",
     "score": 85,
     "score_low": 81,
     "score_high": 88,
     "score_source": "model",
     "score_inputs": [
      "Arena"
     ],
     "usd_month": 20,
     "checked_on": "2026-10-11"
    },
    {
     "category": "research",
     "score": 94,
     "score_low": 87,
     "score_high": 99,
     "score_source": "model",
     "score_inputs": [
      "Arena",
      "Epoch AI"
     ],
     "usd_month": 20,
     "checked_on": "2026-10-11"
    },
    {
     "category": "writing",
     "score": 83,
     "score_low": 78,
     "score_high": 87,
     "score_source": "model",
     "score_inputs": [
      "Arena"
     ],
     "usd_month": 20,
     "checked_on": "2026-10-11"
    }
   ],
   "price": {
    "free_plan": false,
    "usd_month": 20,
    "plans": "ChatGPT Plus $20/month (s9) and Pro include GPT-6.1 Sol; Free and Go do not (s10). At launch it was in ChatGPT Work and Codex, not yet in Chat (s11).",
    "api": "$2 / $10 per 1M input / output tokens, cached input $0.10; long context $4 / $15 (s12)",
    "checked_on": "2026-10-11"
   },
   "how_we_rated": "From arena.ai leaderboards and the Artificial Analysis Intelligence Index, checked 11 Oct 2026. We did not run it ourselves.",
   "sources": [
    {
     "id": "s1",
     "title": "arena.ai Agent leaderboard",
     "url": "https://arena.ai/leaderboard/agent?v=20261011",
     "checked": "2026-10-11"
    },
    {
     "id": "s2",
     "title": "arena.ai Agent Pareto and cost per task",
     "url": "https://arena.ai/leaderboard/agent/pareto?v=20261011",
     "checked": "2026-10-11"
    },
    {
     "id": "s3",
     "title": "Artificial Analysis LLM leaderboard",
     "url": "https://artificialanalysis.ai/leaderboards/models?v=20261011",
     "checked": "2026-10-11"
    },
    {
     "id": "s4",
     "title": "arena.ai Code Arena: WebDev leaderboard",
     "url": "https://arena.ai/leaderboard/code/webdev?v=20261011",
     "checked": "2026-10-11"
    },
    {
     "id": "s5",
     "title": "arena.ai Text: Coding leaderboard",
     "url": "https://arena.ai/leaderboard/chat/text/coding?v=20261011",
     "checked": "2026-10-11"
    },
    {
     "id": "s6",
     "title": "arena.ai Text: Expert leaderboard",
     "url": "https://arena.ai/leaderboard/chat/text/expert?v=20261011",
     "checked": "2026-10-11"
    },
    {
     "id": "s7",
     "title": "arena.ai Text: Creative Writing leaderboard",
     "url": "https://arena.ai/leaderboard/chat/text/creative-writing?v=20261011",
     "checked": "2026-10-11"
    },
    {
     "id": "s8",
     "title": "arena.ai Text: Instruction Following leaderboard",
     "url": "https://arena.ai/leaderboard/chat/text/instruction-following?v=20261011",
     "checked": "2026-10-11"
    },
    {
     "id": "s9",
     "title": "OpenAI Help: What is ChatGPT Plus?",
     "url": "https://help.openai.com/en/articles/6950777-what-is-chatgpt-plus",
     "checked": "2026-10-11"
    },
    {
     "id": "s10",
     "title": "ChatGPT pricing and plan comparison",
     "url": "https://chatgpt.com/pricing",
     "checked": "2026-10-11"
    },
    {
     "id": "s11",
     "title": "OpenAI: Introducing GPT-6.1 Sol",
     "url": "https://openai.com/index/introducing-gpt-6-1-sol/",
     "checked": "2026-10-11"
    },
    {
     "id": "s12",
     "title": "OpenAI API pricing",
     "url": "https://developers.openai.com/api/docs/pricing",
     "checked": "2026-10-11"
    },
    {
     "id": "s13",
     "title": "OpenAI Help: How your data is used to improve model performance",
     "url": "https://help.openai.com/en/articles/5722486-how-your-data-is-used-to-improve-model-performance",
     "checked": "2026-10-11"
    },
    {
     "id": "s14",
     "title": "OpenAI API docs: Your data",
     "url": "https://developers.openai.com/api/docs/guides/your-data",
     "checked": "2026-10-11"
    },
    {
     "id": "s15",
     "title": "arena.ai Agent Alignment Index",
     "url": "https://arena.ai/leaderboard/agent/alignment?v=20261011",
     "checked": "2026-10-11"
    },
    {
     "id": "s16",
     "title": "Unite.AI: OpenAI unveils GPT-6.1 Sol at DevDay",
     "url": "https://www.unite.ai/openai-unveils-gpt-6.1-sol-at-devday-with-new-codex-and-chatgpt-tools/",
     "checked": "2026-10-11"
    }
   ],
   "checked": "2026-10-11"
  },
  {
   "slug": "openai-codex-cloud",
   "url": "https://hayzlabs.com/reviews/openai-codex-cloud/",
   "name": "OpenAI Codex Cloud",
   "maker": "OpenAI",
   "kind": "dev-tool",
   "license": "proprietary",
   "runs": "cloud",
   "released": "2026-09-29",
   "status": "first-look",
   "verdict": "Worth trying",
   "verdict_line": "Codex tasks that keep running in OpenAI's cloud, on Plus and up; no independent test of the cloud side yet.",
   "grades": [],
   "price": {
    "free_plan": false,
    "usd_month": 20,
    "plans": "Plus $20/month; Pro $100, $200 or $500/month; Business $25/user monthly or $20 yearly (s2). Cloud features don't work with an API key alone (s2).",
    "api": null,
    "checked_on": "2026-10-11"
   },
   "how_we_rated": "First look from OpenAI's DevDay recap, Codex pricing and data pages, checked 11 Oct 2026. Not rated yet. We did not run it ourselves.",
   "sources": [
    {
     "id": "s1",
     "title": "OpenAI: DevDay 2026 recap",
     "url": "https://openai.com/index/devday-2026-recap/?v=20261011",
     "checked": "2026-10-11"
    },
    {
     "id": "s2",
     "title": "Codex docs: Pricing",
     "url": "https://learn.chatgpt.com/docs/pricing?v=20261011",
     "checked": "2026-10-11"
    },
    {
     "id": "s3",
     "title": "OpenAI Help: How your data is used to improve model performance",
     "url": "https://help.openai.com/en/articles/5722486-how-your-data-is-used-to-improve-model-performance",
     "checked": "2026-10-11"
    },
    {
     "id": "s4",
     "title": "GitHub Copilot plans",
     "url": "https://github.com/features/copilot/plans?v=20261011",
     "checked": "2026-10-11"
    }
   ],
   "checked": "2026-10-11"
  },
  {
   "slug": "claude-sonnet-5-5",
   "url": "https://hayzlabs.com/reviews/claude-sonnet-5-5/",
   "name": "Claude Sonnet 5.5",
   "maker": "Anthropic",
   "kind": "model",
   "license": "proprietary",
   "runs": "cloud",
   "released": "2026-09-28",
   "status": "graded",
   "verdict": "Use it",
   "verdict_line": "Near-top coding, agents and research for half Opus 5.5's token price; for writing, pick Opus 5.5 or Gemini 3.8 Flash.",
   "grades": [
    {
     "category": "coding",
     "score": 89,
     "score_low": 85,
     "score_high": 92,
     "score_source": "model",
     "score_inputs": [
      "Arena"
     ],
     "usd_month": 0,
     "checked_on": "2026-10-11"
    },
    {
     "category": "agents",
     "score": null,
     "score_low": null,
     "score_high": null,
     "score_source": "model",
     "score_inputs": [
      "Arena"
     ],
     "arena_rank": 5,
     "usd_month": 20,
     "checked_on": "2026-10-11"
    },
    {
     "category": "research",
     "score": 90,
     "score_low": 83,
     "score_high": 94,
     "score_source": "model",
     "score_inputs": [
      "Arena",
      "Epoch AI"
     ],
     "usd_month": 0,
     "checked_on": "2026-10-11"
    },
    {
     "category": "writing",
     "score": 85,
     "score_low": 80,
     "score_high": 89,
     "score_source": "model",
     "score_inputs": [
      "Arena"
     ],
     "usd_month": 0,
     "checked_on": "2026-10-11"
    }
   ],
   "price": {
    "free_plan": true,
    "usd_month": 0,
    "plans": "Free in the Claude app on the Free plan (s9, s15); Pro $20/month for higher limits (s9); API $2 / $10 per 1M tokens (s10).",
    "api": "$2 / $10 per 1M input / output tokens; cache reads $0.10 (s10)",
    "checked_on": "2026-10-11"
   },
   "how_we_rated": "From arena.ai leaderboards and the Artificial Analysis Intelligence Index, checked 11 Oct 2026. We did not run it ourselves.",
   "sources": [
    {
     "id": "s1",
     "title": "arena.ai Code Arena: WebDev leaderboard",
     "url": "https://arena.ai/leaderboard/code/webdev?v=20261011",
     "checked": "2026-10-11"
    },
    {
     "id": "s2",
     "title": "arena.ai Text: Coding leaderboard",
     "url": "https://arena.ai/leaderboard/chat/text/coding?v=20261011",
     "checked": "2026-10-11"
    },
    {
     "id": "s3",
     "title": "arena.ai Agent leaderboard",
     "url": "https://arena.ai/leaderboard/agent?v=20261011",
     "checked": "2026-10-11"
    },
    {
     "id": "s4",
     "title": "arena.ai Agent Pareto and cost per task",
     "url": "https://arena.ai/leaderboard/agent/pareto?v=20261011",
     "checked": "2026-10-11"
    },
    {
     "id": "s5",
     "title": "Artificial Analysis LLM leaderboard",
     "url": "https://artificialanalysis.ai/leaderboards/models?v=20261011",
     "checked": "2026-10-11"
    },
    {
     "id": "s6",
     "title": "arena.ai Text: Expert leaderboard",
     "url": "https://arena.ai/leaderboard/chat/text/expert?v=20261011",
     "checked": "2026-10-11"
    },
    {
     "id": "s7",
     "title": "arena.ai Text: Creative Writing leaderboard",
     "url": "https://arena.ai/leaderboard/chat/text/creative-writing?v=20261011",
     "checked": "2026-10-11"
    },
    {
     "id": "s8",
     "title": "arena.ai Text: Instruction Following leaderboard",
     "url": "https://arena.ai/leaderboard/chat/text/instruction-following?v=20261011",
     "checked": "2026-10-11"
    },
    {
     "id": "s9",
     "title": "Claude plans and pricing",
     "url": "https://claude.com/pricing",
     "checked": "2026-10-11"
    },
    {
     "id": "s10",
     "title": "Claude API pricing",
     "url": "https://platform.claude.com/docs/en/about-claude/pricing",
     "checked": "2026-10-11"
    },
    {
     "id": "s11",
     "title": "Claude privacy: is my data used for model training (consumer)",
     "url": "https://privacy.claude.com/en/articles/10023580-is-my-data-used-for-model-training",
     "checked": "2026-10-11"
    },
    {
     "id": "s12",
     "title": "Claude privacy: is my data used for model training (commercial)",
     "url": "https://privacy.claude.com/en/articles/7996868-is-my-data-used-for-model-training",
     "checked": "2026-10-11"
    },
    {
     "id": "s13",
     "title": "Anthropic: Introducing Claude Sonnet 5.5",
     "url": "https://www.anthropic.com/claude-sonnet-5-5",
     "checked": "2026-10-11"
    },
    {
     "id": "s14",
     "title": "Claude Sonnet 5.5 model overview",
     "url": "https://platform.claude.com/docs/en/models/sonnet-5-5/overview",
     "checked": "2026-10-11"
    },
    {
     "id": "s15",
     "title": "PCWorld: Claude Sonnet 5.5 is here (29 Sep 2026)",
     "url": "https://pcworld.com/article/3246614/claude-sonnet-5-5-is-here-this-prompt-stops-it-from-wasting-your-usage.html",
     "checked": "2026-10-11"
    }
   ],
   "checked": "2026-10-11"
  },
  {
   "slug": "eleven-v4",
   "url": "https://hayzlabs.com/reviews/eleven-v4/",
   "name": "Eleven v4",
   "maker": "ElevenLabs",
   "kind": "model",
   "license": "proprietary",
   "runs": "cloud",
   "released": "2026-09-28",
   "status": "graded",
   "verdict": "Worth trying",
   "verdict_line": "Top-rated expressive voice for audiobooks and characters. Eleven v4 Turbo ties it in blind tests at half the price.",
   "grades": [
    {
     "category": "voice",
     "score": null,
     "score_low": null,
     "score_high": null,
     "score_source": null,
     "usd_month": 0,
     "checked_on": "2026-10-11"
    }
   ],
   "price": {
    "free_plan": true,
    "usd_month": 0,
    "plans": "Free plan with 10k credits a month (no commercial license); Starter $6/month (s2, s3)",
    "api": "$0.08 per 1K characters; the launch promo ended 12 Oct 2026 (s2)",
    "checked_on": "2026-10-19"
   },
   "how_we_rated": "From the Artificial Analysis text to speech arena (blind listener votes), checked 11 Oct 2026. We did not run it ourselves.",
   "sources": [
    {
     "id": "s1",
     "title": "Artificial Analysis text to speech leaderboard",
     "url": "https://artificialanalysis.ai/text-to-speech/leaderboard/provider-voice",
     "checked": "2026-10-11"
    },
    {
     "id": "s2",
     "title": "ElevenLabs API pricing",
     "url": "https://elevenlabs.io/pricing/api?v=1019",
     "checked": "2026-10-19"
    },
    {
     "id": "s3",
     "title": "ElevenLabs pricing",
     "url": "https://elevenlabs.io/pricing",
     "checked": "2026-10-11"
    },
    {
     "id": "s4",
     "title": "ElevenLabs privacy policy",
     "url": "https://elevenlabs.io/privacy-policy",
     "checked": "2026-10-11"
    },
    {
     "id": "s5",
     "title": "ElevenLabs docs: models",
     "url": "https://elevenlabs.io/docs/overview/models",
     "checked": "2026-10-11"
    },
    {
     "id": "s6",
     "title": "ElevenLabs blog: Eleven v4",
     "url": "https://elevenlabs.io/blog/eleven-v4",
     "checked": "2026-10-11"
    }
   ],
   "checked": "2026-10-11"
  },
  {
   "slug": "eleven-v4-turbo",
   "url": "https://hayzlabs.com/reviews/eleven-v4-turbo/",
   "name": "Eleven v4 Turbo",
   "maker": "ElevenLabs",
   "kind": "model",
   "license": "proprietary",
   "runs": "cloud",
   "released": "2026-09-28",
   "status": "graded",
   "verdict": "Use it",
   "verdict_line": "Top-rated natural voice in blind tests at half Eleven v4's price, fast enough for live agents. Not the cheapest.",
   "grades": [
    {
     "category": "voice",
     "score": null,
     "score_low": null,
     "score_high": null,
     "score_source": null,
     "usd_month": 0,
     "checked_on": "2026-10-11"
    }
   ],
   "price": {
    "free_plan": true,
    "usd_month": 0,
    "plans": "Free plan with 10k credits a month (no commercial license); Starter $6/month (s2, s3)",
    "api": "$0.04 per 1K characters; the launch promo ended 12 Oct 2026 (s2)",
    "checked_on": "2026-10-19"
   },
   "how_we_rated": "From the Artificial Analysis text to speech arena (blind listener votes), checked 11 Oct 2026. We did not run it ourselves.",
   "sources": [
    {
     "id": "s1",
     "title": "Artificial Analysis text to speech leaderboard",
     "url": "https://artificialanalysis.ai/text-to-speech/leaderboard/provider-voice",
     "checked": "2026-10-11"
    },
    {
     "id": "s2",
     "title": "ElevenLabs API pricing",
     "url": "https://elevenlabs.io/pricing/api?v=1019",
     "checked": "2026-10-19"
    },
    {
     "id": "s3",
     "title": "ElevenLabs pricing",
     "url": "https://elevenlabs.io/pricing",
     "checked": "2026-10-11"
    },
    {
     "id": "s4",
     "title": "ElevenLabs privacy policy",
     "url": "https://elevenlabs.io/privacy-policy",
     "checked": "2026-10-11"
    },
    {
     "id": "s5",
     "title": "ElevenLabs docs: models",
     "url": "https://elevenlabs.io/docs/overview/models",
     "checked": "2026-10-11"
    },
    {
     "id": "s6",
     "title": "ElevenLabs blog: Eleven v4",
     "url": "https://elevenlabs.io/blog/eleven-v4",
     "checked": "2026-10-11"
    }
   ],
   "checked": "2026-10-11"
  },
  {
   "slug": "claude-opus-5-5",
   "url": "https://hayzlabs.com/reviews/claude-opus-5-5/",
   "name": "Claude Opus 5.5",
   "maker": "Anthropic",
   "kind": "model",
   "license": "proprietary",
   "runs": "cloud",
   "released": "2026-09-22",
   "status": "graded",
   "verdict": "Use it",
   "verdict_line": "Top public pick for coding, agents and hard questions at a mid price; skip it for bulk work where Sonnet 5.5 is enough.",
   "grades": [
    {
     "category": "coding",
     "score": 100,
     "score_low": 89,
     "score_high": 100,
     "score_source": "model",
     "score_inputs": [
      "Arena",
      "Epoch AI"
     ],
     "usd_month": 20,
     "checked_on": "2026-10-11"
    },
    {
     "category": "agents",
     "score": null,
     "score_low": null,
     "score_high": null,
     "score_source": "model",
     "score_inputs": [
      "Arena"
     ],
     "arena_rank": 1,
     "usd_month": 20,
     "checked_on": "2026-10-11"
    },
    {
     "category": "research",
     "score": 94,
     "score_low": 88,
     "score_high": 99,
     "score_source": "model",
     "score_inputs": [
      "Arena",
      "Epoch AI"
     ],
     "usd_month": 20,
     "checked_on": "2026-10-11"
    },
    {
     "category": "writing",
     "score": 100,
     "score_low": 96,
     "score_high": 100,
     "score_source": "model",
     "score_inputs": [
      "Arena"
     ],
     "usd_month": 20,
     "checked_on": "2026-10-11"
    }
   ],
   "price": {
    "free_plan": false,
    "usd_month": 20,
    "plans": "Pro $20/month ($17/month billed yearly); Max from $100/month. Free plan does not include Opus (s9).",
    "api": "$4 / $20 per 1M input / output tokens; fast mode (research preview) $8 / $40 (s10)",
    "checked_on": "2026-10-11"
   },
   "how_we_rated": "From arena.ai leaderboards and the Artificial Analysis Intelligence Index, checked 11 Oct 2026. We did not run it ourselves.",
   "sources": [
    {
     "id": "s1",
     "title": "arena.ai Code Arena: WebDev leaderboard",
     "url": "https://arena.ai/leaderboard/code/webdev?v=20261011",
     "checked": "2026-10-11"
    },
    {
     "id": "s2",
     "title": "arena.ai Text: Coding leaderboard",
     "url": "https://arena.ai/leaderboard/chat/text/coding?v=20261011",
     "checked": "2026-10-11"
    },
    {
     "id": "s3",
     "title": "arena.ai Agent leaderboard",
     "url": "https://arena.ai/leaderboard/agent?v=20261011",
     "checked": "2026-10-11"
    },
    {
     "id": "s4",
     "title": "arena.ai Agent Pareto and cost per task",
     "url": "https://arena.ai/leaderboard/agent/pareto?v=20261011",
     "checked": "2026-10-11"
    },
    {
     "id": "s5",
     "title": "Artificial Analysis LLM leaderboard",
     "url": "https://artificialanalysis.ai/leaderboards/models?v=20261011",
     "checked": "2026-10-11"
    },
    {
     "id": "s6",
     "title": "arena.ai Text: Expert leaderboard",
     "url": "https://arena.ai/leaderboard/chat/text/expert?v=20261011",
     "checked": "2026-10-11"
    },
    {
     "id": "s7",
     "title": "arena.ai Text: Creative Writing leaderboard",
     "url": "https://arena.ai/leaderboard/chat/text/creative-writing?v=20261011",
     "checked": "2026-10-11"
    },
    {
     "id": "s8",
     "title": "arena.ai Text: Instruction Following leaderboard",
     "url": "https://arena.ai/leaderboard/chat/text/instruction-following?v=20261011",
     "checked": "2026-10-11"
    },
    {
     "id": "s9",
     "title": "Claude plans and pricing",
     "url": "https://claude.com/pricing",
     "checked": "2026-10-11"
    },
    {
     "id": "s10",
     "title": "Claude API pricing",
     "url": "https://platform.claude.com/docs/en/about-claude/pricing",
     "checked": "2026-10-11"
    },
    {
     "id": "s11",
     "title": "Claude privacy: is my data used for model training (consumer)",
     "url": "https://privacy.claude.com/en/articles/10023580-is-my-data-used-for-model-training",
     "checked": "2026-10-11"
    },
    {
     "id": "s12",
     "title": "Claude privacy: is my data used for model training (commercial)",
     "url": "https://privacy.claude.com/en/articles/7996868-is-my-data-used-for-model-training",
     "checked": "2026-10-11"
    },
    {
     "id": "s13",
     "title": "Anthropic: Introducing Claude Opus 5.5",
     "url": "https://www.anthropic.com/claude-opus-5-5",
     "checked": "2026-10-11"
    },
    {
     "id": "s14",
     "title": "Claude models overview",
     "url": "https://platform.claude.com/docs/en/models/overview",
     "checked": "2026-10-11"
    }
   ],
   "checked": "2026-10-11"
  },
  {
   "slug": "gemini-3-8-flash-tts",
   "url": "https://hayzlabs.com/reviews/gemini-3-8-flash-tts/",
   "name": "Gemini 3.8 Flash TTS",
   "maker": "Google",
   "kind": "model",
   "license": "proprietary",
   "runs": "cloud",
   "released": "2026-09-22",
   "status": "graded",
   "verdict": "Use it",
   "verdict_line": "Near-top AI voices for less than half the price of ElevenLabs, with a free tier. Prices double on 1 Jan 2027.",
   "grades": [
    {
     "category": "voice",
     "score": null,
     "score_low": null,
     "score_high": null,
     "score_source": null,
     "usd_month": 0,
     "checked_on": "2026-10-11"
    }
   ],
   "price": {
    "free_plan": true,
    "usd_month": 0,
    "plans": "Free tier in the Gemini API: input and output free of charge on Standard (s2)",
    "api": "$0.50 per 1M text input tokens and $9.00 per 1M audio output tokens through 31 Dec 2026; $1.00 / $18.00 from 1 Jan 2027; Batch half price (s2)",
    "checked_on": "2026-10-11"
   },
   "how_we_rated": "From the Artificial Analysis text to speech arena (blind listener votes), checked 11 Oct 2026. We did not run it ourselves.",
   "sources": [
    {
     "id": "s1",
     "title": "Artificial Analysis text to speech leaderboard",
     "url": "https://artificialanalysis.ai/text-to-speech/leaderboard/provider-voice",
     "checked": "2026-10-11"
    },
    {
     "id": "s2",
     "title": "Gemini API pricing",
     "url": "https://ai.google.dev/gemini-api/docs/pricing?v=20261011",
     "checked": "2026-10-11"
    },
    {
     "id": "s3",
     "title": "Gemini API docs: speech generation",
     "url": "https://ai.google.dev/gemini-api/docs/speech-generation?v=20261011",
     "checked": "2026-10-11"
    },
    {
     "id": "s4",
     "title": "Gemini API changelog",
     "url": "https://ai.google.dev/gemini-api/docs/changelog?v=20261011",
     "checked": "2026-10-11"
    }
   ],
   "checked": "2026-10-11"
  },
  {
   "slug": "gemini-3-8-flash-lite-tts",
   "url": "https://hayzlabs.com/reviews/gemini-3-8-flash-lite-tts/",
   "name": "Gemini 3.8 Flash-Lite TTS",
   "maker": "Google",
   "kind": "model",
   "license": "proprietary",
   "runs": "cloud",
   "released": "2026-09-22",
   "status": "graded",
   "verdict": "Use it",
   "verdict_line": "Top-ten text to speech for a third less than Gemini 3.8 Flash TTS, with a free tier; pick Eleven v4 for the best voices.",
   "grades": [
    {
     "category": "voice",
     "score": null,
     "score_low": null,
     "score_high": null,
     "score_source": null,
     "usd_month": 0,
     "checked_on": "2026-10-11"
    }
   ],
   "price": {
    "free_plan": true,
    "usd_month": 0,
    "plans": "Free tier in the Gemini API: input and output free of charge on Standard (s2)",
    "api": "$0.50 per 1M text input tokens and $6.00 per 1M audio output tokens through 31 Dec 2026; $1.00 / $12.00 from 1 Jan 2027; Batch half price (s2)",
    "checked_on": "2026-10-11"
   },
   "how_we_rated": "From the Artificial Analysis text to speech arena (blind listener votes), checked 11 Oct 2026. We did not run it ourselves.",
   "sources": [
    {
     "id": "s1",
     "title": "Artificial Analysis text to speech leaderboard",
     "url": "https://artificialanalysis.ai/text-to-speech/leaderboard/provider-voice",
     "checked": "2026-10-11"
    },
    {
     "id": "s2",
     "title": "Gemini API pricing",
     "url": "https://ai.google.dev/gemini-api/docs/pricing?v=20261011",
     "checked": "2026-10-11"
    },
    {
     "id": "s3",
     "title": "Gemini API changelog",
     "url": "https://ai.google.dev/gemini-api/docs/changelog?v=20261011",
     "checked": "2026-10-11"
    }
   ],
   "checked": "2026-10-11"
  },
  {
   "slug": "grok-4-7",
   "url": "https://hayzlabs.com/reviews/grok-4-7/",
   "name": "Grok 4.7",
   "maker": "xAI",
   "kind": "model",
   "license": "proprietary",
   "runs": "cloud",
   "released": "2026-09-21",
   "status": "graded",
   "verdict": "Skip",
   "verdict_line": "Decent agent and web-app work, weak at writing; GPT-6.1 Sol ranks higher for less per task.",
   "grades": [
    {
     "category": "agents",
     "score": null,
     "score_low": null,
     "score_high": null,
     "score_source": "model",
     "score_inputs": [
      "Arena"
     ],
     "arena_rank": 16,
     "usd_month": 30,
     "checked_on": "2026-10-11"
    },
    {
     "category": "coding",
     "score": 55,
     "score_low": 52,
     "score_high": 57,
     "score_source": "model",
     "score_inputs": [
      "Arena"
     ],
     "usd_month": 30,
     "checked_on": "2026-10-11"
    },
    {
     "category": "writing",
     "score": 73,
     "score_low": 70,
     "score_high": 77,
     "score_source": "model",
     "score_inputs": [
      "Arena"
     ],
     "usd_month": 30,
     "checked_on": "2026-10-11"
    }
   ],
   "price": {
    "free_plan": false,
    "usd_month": 30,
    "plans": "xAI's plan page names no model on any plan; Premium ($30/month) promises the \"latest Grok models\" (s7).",
    "api": "$2 / $6 per 1M input / output tokens, 500K context (s8); a fast variant costs twice as much (s9)",
    "checked_on": "2026-10-18"
   },
   "how_we_rated": "From arena.ai leaderboards and the Artificial Analysis Intelligence Index, checked 11 Oct 2026. We did not run it ourselves.",
   "sources": [
    {
     "id": "s1",
     "title": "arena.ai Agent leaderboard",
     "url": "https://arena.ai/leaderboard/agent?v=20261011",
     "checked": "2026-10-11"
    },
    {
     "id": "s2",
     "title": "Artificial Analysis LLM leaderboard",
     "url": "https://artificialanalysis.ai/leaderboards/models?v=20261011",
     "checked": "2026-10-11"
    },
    {
     "id": "s3",
     "title": "arena.ai Code Arena: WebDev leaderboard",
     "url": "https://arena.ai/leaderboard/code/webdev?v=20261011",
     "checked": "2026-10-11"
    },
    {
     "id": "s4",
     "title": "arena.ai Text: Coding leaderboard",
     "url": "https://arena.ai/leaderboard/chat/text/coding?v=20261011",
     "checked": "2026-10-11"
    },
    {
     "id": "s5",
     "title": "arena.ai Text: Creative Writing leaderboard",
     "url": "https://arena.ai/leaderboard/chat/text/creative-writing?v=20261011",
     "checked": "2026-10-11"
    },
    {
     "id": "s6",
     "title": "arena.ai Text: Instruction Following leaderboard",
     "url": "https://arena.ai/leaderboard/chat/text/instruction-following?v=20261011",
     "checked": "2026-10-11"
    },
    {
     "id": "s7",
     "title": "xAI pricing (Grok plans)",
     "url": "https://x.ai/pricing",
     "checked": "2026-10-18"
    },
    {
     "id": "s8",
     "title": "xAI API docs: Models and pricing",
     "url": "https://docs.x.ai/developers/models",
     "checked": "2026-10-11"
    },
    {
     "id": "s9",
     "title": "xAI: Grok 4.7",
     "url": "https://x.ai/news/grok-4-7",
     "checked": "2026-10-11"
    },
    {
     "id": "s10",
     "title": "xAI Consumer FAQs",
     "url": "https://x.ai/legal/faq",
     "checked": "2026-10-11"
    },
    {
     "id": "s11",
     "title": "arena.ai Agent Alignment Index",
     "url": "https://arena.ai/leaderboard/agent/alignment?v=20261011",
     "checked": "2026-10-11"
    }
   ],
   "checked": "2026-10-11"
  },
  {
   "slug": "mimo-v2-6-pro",
   "url": "https://hayzlabs.com/reviews/mimo-v2-6-pro/",
   "name": "MiMo-V2.6-Pro",
   "maker": "Xiaomi",
   "kind": "open-source",
   "license": "open-source",
   "runs": "cloud",
   "released": "2026-09-21",
   "status": "graded",
   "verdict": "Worth trying",
   "verdict_line": "Top open-weights model on the Artificial Analysis index, at cents per task; closed leaders still rank well above it.",
   "grades": [
    {
     "category": "coding",
     "score": 52,
     "score_low": 50,
     "score_high": 54,
     "score_source": "model",
     "score_inputs": [
      "Arena"
     ],
     "usd_month": 0,
     "checked_on": "2026-10-11"
    },
    {
     "category": "agents",
     "score": null,
     "score_low": null,
     "score_high": null,
     "score_source": "model",
     "score_inputs": [
      "Arena"
     ],
     "arena_rank": 13,
     "usd_month": 0,
     "checked_on": "2026-10-11"
    },
    {
     "category": "research",
     "score": 94,
     "score_low": 86,
     "score_high": 100,
     "score_source": "model",
     "score_inputs": [
      "Arena"
     ],
     "usd_month": 0,
     "checked_on": "2026-10-11"
    }
   ],
   "price": {
    "free_plan": false,
    "usd_month": 0,
    "plans": "Try it in Xiaomi MiMo AI Studio or MiMo Desktop, which do not state a price (s9). Free weights under MIT. On Xiaomi's API, $0.435 input and $0.87 output per 1M tokens (s9).",
    "api": null,
    "checked_on": "2026-10-11"
   },
   "how_we_rated": "From arena.ai leaderboards and the Artificial Analysis Intelligence Index, checked 11 Oct 2026. We did not run it ourselves.",
   "sources": [
    {
     "id": "s1",
     "title": "arena.ai Code Arena: WebDev leaderboard",
     "url": "https://arena.ai/leaderboard/code/webdev?v=20261011",
     "checked": "2026-10-11"
    },
    {
     "id": "s2",
     "title": "arena.ai Text: Coding leaderboard",
     "url": "https://arena.ai/leaderboard/chat/text/coding?v=20261011",
     "checked": "2026-10-11"
    },
    {
     "id": "s3",
     "title": "arena.ai Agent leaderboard",
     "url": "https://arena.ai/leaderboard/agent?v=20261011",
     "checked": "2026-10-11"
    },
    {
     "id": "s4",
     "title": "arena.ai Agent Pareto and cost per task",
     "url": "https://arena.ai/leaderboard/agent/pareto?v=20261011",
     "checked": "2026-10-11"
    },
    {
     "id": "s5",
     "title": "Artificial Analysis LLM leaderboard",
     "url": "https://artificialanalysis.ai/leaderboards/models?v=20261011",
     "checked": "2026-10-11"
    },
    {
     "id": "s6",
     "title": "arena.ai Text: Expert leaderboard",
     "url": "https://arena.ai/leaderboard/chat/text/expert?v=20261011",
     "checked": "2026-10-11"
    },
    {
     "id": "s7",
     "title": "Xiaomi MiMo Platform privacy policy",
     "url": "https://privacy.mi.com/XiaomiMiMoPlatform/en_GB/",
     "checked": "2026-10-11"
    },
    {
     "id": "s9",
     "title": "Xiaomi MiMo-V2.6 launch post (prices)",
     "url": "https://mimo.xiaomi.com/mimo-v2-6/article.html",
     "checked": "2026-10-11"
    },
    {
     "id": "s8",
     "title": "MiMo-V2.6-Pro-RL model card (Hugging Face)",
     "url": "https://huggingface.co/XiaomiMiMo/MiMo-V2.6-Pro-RL",
     "checked": "2026-10-11"
    }
   ],
   "checked": "2026-10-11"
  },
  {
   "slug": "qwen-image-2-1",
   "url": "https://hayzlabs.com/reviews/qwen-image-2-1/",
   "name": "Qwen-Image-2.1",
   "maker": "Alibaba",
   "kind": "open-source",
   "license": "open-weights",
   "runs": "local",
   "released": "2026-09-20",
   "status": "graded",
   "verdict": "Depends",
   "verdict_line": "Best scoring image model you can download and run yourself, but its license bars commercial use. Fine for research.",
   "grades": [
    {
     "category": "images",
     "score": 48,
     "score_low": 46,
     "score_high": 49,
     "score_source": "model",
     "score_inputs": [
      "Arena"
     ],
     "usd_month": 0,
     "checked_on": "2026-10-11"
    }
   ],
   "price": {
    "free_plan": true,
    "usd_month": 0,
    "plans": "Free download; you supply the hardware. No hosted API listed on Artificial Analysis (s3).",
    "api": "None from Qwen listed on Artificial Analysis (s3)",
    "checked_on": "2026-10-11"
   },
   "how_we_rated": "From the arena.ai and Artificial Analysis image leaderboards and Qwen's model page and license, checked 11 Oct 2026. We did not run it ourselves.",
   "sources": [
    {
     "id": "s1",
     "title": "arena.ai Text to Image leaderboard",
     "url": "https://arena.ai/leaderboard/image/text-to-image?v=20261011",
     "checked": "2026-10-11"
    },
    {
     "id": "s2",
     "title": "arena.ai Image Edit leaderboard (Single Image Edit)",
     "url": "https://arena.ai/leaderboard/image/image-edit",
     "checked": "2026-10-11"
    },
    {
     "id": "s3",
     "title": "Artificial Analysis text-to-image leaderboard",
     "url": "https://artificialanalysis.ai/image/leaderboard/text-to-image?v=20261011",
     "checked": "2026-10-11"
    },
    {
     "id": "s4",
     "title": "Artificial Analysis image editing leaderboard",
     "url": "https://artificialanalysis.ai/image/leaderboard/editing?v=20261011",
     "checked": "2026-10-11"
    },
    {
     "id": "s5",
     "title": "Qwen-Image-2.1 model page on Hugging Face",
     "url": "https://huggingface.co/Qwen/Qwen-Image-2.1",
     "checked": "2026-10-11"
    },
    {
     "id": "s6",
     "title": "Qwen Research License Agreement (Qwen-Image-2.1)",
     "url": "https://huggingface.co/Qwen/Qwen-Image-2.1/blob/main/LICENSE",
     "checked": "2026-10-11"
    }
   ],
   "checked": "2026-10-11"
  },
  {
   "slug": "wispr-flow",
   "url": "https://hayzlabs.com/reviews/wispr-flow/",
   "name": "Wispr Flow",
   "maker": "Wispr",
   "kind": "app",
   "license": "proprietary",
   "runs": "cloud",
   "released": "2026-09-17",
   "status": "first-look",
   "verdict": "Worth trying",
   "verdict_line": "Popular voice dictation that types cleaned-up text in any app; free for 2,000 words a week, Pro $15/month.",
   "grades": [],
   "price": {
    "free_plan": true,
    "usd_month": 0,
    "plans": "Free: 2,000 words/week on desktop, 1,000 on mobile. Pro $15/user/month, or $12 billed yearly, unlimited words. Growth and Enterprise for teams (s1).",
    "api": null,
    "checked_on": "2026-10-11"
   },
   "how_we_rated": "First look from Wispr's home page, pricing page, privacy page and update notes, checked 11 Oct 2026. Not rated yet. We did not run it ourselves.",
   "sources": [
    {
     "id": "s1",
     "title": "Wispr Flow pricing",
     "url": "https://wisprflow.ai/pricing?v=20261011",
     "checked": "2026-10-11"
    },
    {
     "id": "s2",
     "title": "Wispr Flow privacy",
     "url": "https://wisprflow.ai/privacy?v=20261011",
     "checked": "2026-10-11"
    },
    {
     "id": "s3",
     "title": "Wispr Flow home page",
     "url": "https://wisprflow.ai/?v=20261011",
     "checked": "2026-10-11"
    },
    {
     "id": "s4",
     "title": "Wispr Flow: What's new",
     "url": "https://wisprflow.ai/whats-new?v=20261011",
     "checked": "2026-10-11"
    }
   ],
   "checked": "2026-10-11"
  },
  {
   "slug": "claude",
   "url": "https://hayzlabs.com/reviews/claude/",
   "name": "Claude",
   "maker": "Anthropic",
   "kind": "app",
   "license": "proprietary",
   "runs": "cloud",
   "released": "2026-09-16",
   "status": "graded",
   "verdict": "Worth trying",
   "verdict_line": "Free runs Sonnet 5.5, near the top for coding and hard questions. Pay for Pro for agents and for writing with Opus 5.5.",
   "grades": [
    {
     "category": "coding",
     "score": 80,
     "score_low": 80,
     "score_high": 80,
     "score_source": "checklist",
     "usd_month": 0,
     "checked_on": "2026-11-02"
    },
    {
     "category": "research",
     "score": 80,
     "score_low": 80,
     "score_high": 80,
     "score_source": "checklist",
     "usd_month": 0,
     "checked_on": "2026-11-02"
    },
    {
     "category": "agents",
     "score": 70,
     "score_low": 70,
     "score_high": 70,
     "score_source": "checklist",
     "usd_month": 20,
     "checked_on": "2026-11-02"
    },
    {
     "category": "writing",
     "score": 70,
     "score_low": 70,
     "score_high": 70,
     "score_source": "checklist",
     "usd_month": 20,
     "checked_on": "2026-11-02"
    }
   ],
   "price": {
    "free_plan": true,
    "usd_month": 0,
    "plans": "Free; Pro $20/month or $17/month billed yearly; Max $100/month; Ultra $200/month; Team from $20/seat/month billed yearly (s5)",
    "api": null,
    "checked_on": "2026-11-02"
   },
   "how_we_rated": "From the SPEC checklist, scored per plan with one rule for every app: each plan is scored on the best model it lets you choose, as documented on the maker's own plan and help pages, and we name it; a default the maker names does not lower the score. Does the job is 2 when that model sits in the top band of Arena's ranks for the job, else 1; underlying quality is 2 for the top band, 1 for the next band, 0 below. The score comes from the higher-scoring plan; a tie goes to the cheaper one. Ranks are from arena.ai, and plans and privacy from Anthropic's own pages, checked 11 Oct 2026. We did not run it ourselves. Where we point to an Anthropic product, we also checked our reviews for the best non-Anthropic option and name it. This product is made by Anthropic, whose Claude models help draft our reviews.",
   "sources": [
    {
     "id": "s1",
     "title": "arena.ai Code Arena: WebDev leaderboard",
     "url": "https://arena.ai/leaderboard/code/webdev?v=20261011",
     "checked": "2026-10-11"
    },
    {
     "id": "s2",
     "title": "arena.ai Text: Expert leaderboard",
     "url": "https://arena.ai/leaderboard/chat/text/expert?v=20261011",
     "checked": "2026-10-11"
    },
    {
     "id": "s3",
     "title": "arena.ai Agent leaderboard",
     "url": "https://arena.ai/leaderboard/agent?v=20261011",
     "checked": "2026-10-11"
    },
    {
     "id": "s4",
     "title": "arena.ai Text: Creative Writing leaderboard",
     "url": "https://arena.ai/leaderboard/chat/text/creative-writing?v=20261011",
     "checked": "2026-10-11"
    },
    {
     "id": "s5",
     "title": "Claude plans and pricing",
     "url": "https://claude.com/pricing?v=20261102",
     "checked": "2026-11-02"
    },
    {
     "id": "s6",
     "title": "PCWorld: Claude Sonnet 5.5 is here (29 Sep 2026)",
     "url": "https://pcworld.com/article/3246614/claude-sonnet-5-5-is-here-this-prompt-stops-it-from-wasting-your-usage.html",
     "checked": "2026-10-11"
    },
    {
     "id": "s7",
     "title": "Claude release notes",
     "url": "https://support.claude.com/en/articles/12138966-release-notes",
     "checked": "2026-10-11"
    },
    {
     "id": "s8",
     "title": "Claude privacy: is my data used for model training (consumer)",
     "url": "https://privacy.claude.com/en/articles/10023580-is-my-data-used-for-model-training",
     "checked": "2026-10-11"
    },
    {
     "id": "s9",
     "title": "Anthropic: Updates to our consumer terms",
     "url": "https://www.anthropic.com/news/updates-to-our-consumer-terms",
     "checked": "2026-10-11"
    },
    {
     "id": "s10",
     "title": "Tom's Guide: ChatGPT-6 vs Claude Opus 5.5 with 5 everyday prompts (24 Sep 2026)",
     "url": "https://www.tomsguide.com/ai/i-tested-chatgpt-6-vs-claude-opus-5-5-with-5-everyday-prompts-it-wasnt-even-close",
     "checked": "2026-10-11"
    },
    {
     "id": "s11",
     "title": "Anthropic: Introducing Claude Sonnet 5.5 (28 Sep 2026)",
     "url": "https://www.anthropic.com/claude-sonnet-5-5",
     "checked": "2026-10-11"
    }
   ],
   "checked": "2026-10-11"
  },
  {
   "slug": "gemini-3-8-live",
   "url": "https://hayzlabs.com/reviews/gemini-3-8-live/",
   "name": "Gemini 3.8 Live",
   "maker": "Google",
   "kind": "model",
   "license": "proprietary",
   "runs": "cloud",
   "released": "2026-09-15",
   "status": "graded",
   "verdict": "Use it",
   "verdict_line": "Among the best-liked voice agent models, free to start and the cheapest of the top group. Weak at hard support tasks.",
   "grades": [
    {
     "category": "voice",
     "score": null,
     "score_low": null,
     "score_high": null,
     "score_source": null,
     "usd_month": 0,
     "checked_on": "2026-10-11"
    }
   ],
   "price": {
    "free_plan": true,
    "usd_month": 0,
    "plans": "Free tier in the Gemini API: input and output free of charge (s3)",
    "api": "Audio $3.00 per 1M input tokens (about $0.005/min), $12.00 per 1M output tokens (about $0.018/min); text $0.75 / $4.50; Extended Thinking same price (s3)",
    "checked_on": "2026-10-11"
   },
   "how_we_rated": "From the Artificial Analysis Speech Agent Arena (blind votes) and Speech to Speech Index, checked 11 Oct 2026. We did not run it ourselves.",
   "sources": [
    {
     "id": "s1",
     "title": "Artificial Analysis Speech Agent Arena leaderboard",
     "url": "https://artificialanalysis.ai/speech-to-speech/arena?v=20261011",
     "checked": "2026-10-11"
    },
    {
     "id": "s2",
     "title": "Artificial Analysis speech to speech leaderboard",
     "url": "https://artificialanalysis.ai/speech-to-speech?v=20261011",
     "checked": "2026-10-11"
    },
    {
     "id": "s3",
     "title": "Gemini API pricing",
     "url": "https://ai.google.dev/gemini-api/docs/pricing?v=20261011",
     "checked": "2026-10-11"
    },
    {
     "id": "s4",
     "title": "Gemini API changelog",
     "url": "https://ai.google.dev/gemini-api/docs/changelog?v=20261011",
     "checked": "2026-10-11"
    }
   ],
   "checked": "2026-10-11"
  },
  {
   "slug": "suno-v6",
   "url": "https://hayzlabs.com/reviews/suno-v6/",
   "name": "Suno v6",
   "maker": "Suno",
   "kind": "model",
   "license": "proprietary",
   "runs": "cloud",
   "released": "2026-09-09",
   "status": "first-look",
   "verdict": "Worth trying",
   "verdict_line": "Suno's new song models; v6-mini is free, but free songs can't be downloaded and your uploads can train it.",
   "grades": [],
   "price": {
    "free_plan": true,
    "usd_month": 8,
    "plans": "Free: 50 credits a day, v6-mini only, no commercial rights, no monthly song downloads; Pro $8/month: 2,500 credits, v6 and v6-wild, commercial rights, 20 downloads a month; Premier $24/month: 10,000 credits, 60 downloads (s2).",
    "api": null,
    "checked_on": "2026-10-11"
   },
   "how_we_rated": "First look from Suno's launch post, pricing and privacy pages and launch coverage, checked 11 Oct 2026. Not rated yet. We did not run it ourselves.",
   "sources": [
    {
     "id": "s1",
     "title": "Suno blog: Introducing v6 (9 Sep 2026)",
     "url": "https://suno.com/blog/introducing-v6?v=20261011",
     "checked": "2026-10-11"
    },
    {
     "id": "s2",
     "title": "Suno pricing",
     "url": "https://suno.com/pricing?v=20261011",
     "checked": "2026-10-11"
    },
    {
     "id": "s3",
     "title": "Suno privacy notice (last modified 23 Jun 2026)",
     "url": "https://suno.com/privacy?v=20261011",
     "checked": "2026-10-11"
    },
    {
     "id": "s4",
     "title": "Engadget: Suno trained its v6 AI music models with help from Warner and BMG (9 Sep 2026)",
     "url": "https://engadget.com/2251539/suno-trained-its-v6-ai-music-models-with-help-from-warner-and-bmg/",
     "checked": "2026-10-11"
    },
    {
     "id": "s5",
     "title": "MusicRadar: Suno has rebuilt its AI music models from scratch with licensed music (9 Sep 2026)",
     "url": "https://www.musicradar.com/music-tech/suno-has-rebuilt-its-ai-music-models-from-scratch-with-licensed-music",
     "checked": "2026-10-11"
    },
    {
     "id": "s6",
     "title": "Artificial Analysis music leaderboard: vocals (AA-Music-Vocal v1.1)",
     "url": "https://artificialanalysis.ai/music/leaderboard/vocals?v=20261011",
     "checked": "2026-10-11"
    }
   ],
   "checked": "2026-10-11"
  },
  {
   "slug": "gpt-image-2-5-flare",
   "url": "https://hayzlabs.com/reviews/gpt-image-2-5-flare/",
   "name": "GPT Image 2.5 Flare",
   "maker": "OpenAI",
   "kind": "model",
   "license": "proprietary",
   "runs": "cloud",
   "released": "2026-09-08",
   "status": "graded",
   "verdict": "Depends",
   "verdict_line": "Second on every image board, at Sunburst's price; OpenAI says it is faster. If waits are fine, Sunburst scores higher.",
   "grades": [
    {
     "category": "images",
     "score": 92,
     "score_low": 90,
     "score_high": 94,
     "score_source": "model",
     "score_inputs": [
      "Arena"
     ],
     "usd_month": null,
     "checked_on": "2026-10-11"
    }
   ],
   "price": {
    "free_plan": false,
    "usd_month": null,
    "plans": "API only under this name. ChatGPT Images 2.5 is offered to all ChatGPT users, but OpenAI does not say which 2.5 model ChatGPT uses (s6).",
    "api": "$5 text in, $8 image in, $30 image out per 1M tokens, the same as Sunburst; Batch halves the image rates (s5)",
    "checked_on": "2026-10-11"
   },
   "how_we_rated": "From the arena.ai and Artificial Analysis image leaderboards, checked 11 Oct 2026. We did not run it ourselves.",
   "sources": [
    {
     "id": "s1",
     "title": "arena.ai Text to Image leaderboard",
     "url": "https://arena.ai/leaderboard/image/text-to-image?v=20261011",
     "checked": "2026-10-11"
    },
    {
     "id": "s2",
     "title": "arena.ai Image Edit leaderboard (Single Image Edit)",
     "url": "https://arena.ai/leaderboard/image/image-edit",
     "checked": "2026-10-11"
    },
    {
     "id": "s3",
     "title": "Artificial Analysis text-to-image leaderboard",
     "url": "https://artificialanalysis.ai/image/leaderboard/text-to-image?v=20261011",
     "checked": "2026-10-11"
    },
    {
     "id": "s4",
     "title": "Artificial Analysis image editing leaderboard",
     "url": "https://artificialanalysis.ai/image/leaderboard/editing?v=20261011",
     "checked": "2026-10-11"
    },
    {
     "id": "s5",
     "title": "OpenAI API pricing",
     "url": "https://developers.openai.com/api/docs/pricing?v=20261011",
     "checked": "2026-10-11"
    },
    {
     "id": "s6",
     "title": "OpenAI: Introducing ChatGPT Images 2.5",
     "url": "https://openai.com/index/introducing-chatgpt-images-2-5/",
     "checked": "2026-10-11"
    },
    {
     "id": "s7",
     "title": "OpenAI enterprise privacy",
     "url": "https://openai.com/enterprise-privacy/",
     "checked": "2026-10-11"
    }
   ],
   "checked": "2026-10-11"
  },
  {
   "slug": "gpt-image-2-5-sunburst",
   "url": "https://hayzlabs.com/reviews/gpt-image-2-5-sunburst/",
   "name": "GPT Image 2.5 Sunburst",
   "maker": "OpenAI",
   "kind": "model",
   "license": "proprietary",
   "runs": "cloud",
   "released": "2026-09-08",
   "status": "graded",
   "verdict": "Worth trying",
   "verdict_line": "Top image model for quality and precise edits, about 21 cents a max-quality image. On a budget, use Nano Banana 2.1.",
   "grades": [
    {
     "category": "images",
     "score": 100,
     "score_low": 98,
     "score_high": 100,
     "score_source": "model",
     "score_inputs": [
      "Arena"
     ],
     "usd_month": null,
     "checked_on": "2026-10-11"
    }
   ],
   "price": {
    "free_plan": false,
    "usd_month": null,
    "plans": "API only under this name. ChatGPT Images 2.5 is offered to all ChatGPT users, but OpenAI does not say which 2.5 model ChatGPT uses (s6).",
    "api": "$5 text in, $8 image in, $30 image out per 1M tokens; Batch halves the image rates (s5)",
    "checked_on": "2026-10-11"
   },
   "how_we_rated": "From the arena.ai and Artificial Analysis image leaderboards, checked 11 Oct 2026. We did not run it ourselves.",
   "sources": [
    {
     "id": "s1",
     "title": "arena.ai Text to Image leaderboard",
     "url": "https://arena.ai/leaderboard/image/text-to-image?v=20261011",
     "checked": "2026-10-11"
    },
    {
     "id": "s2",
     "title": "arena.ai Image Edit leaderboard (Single Image Edit)",
     "url": "https://arena.ai/leaderboard/image/image-edit",
     "checked": "2026-10-11"
    },
    {
     "id": "s3",
     "title": "Artificial Analysis text-to-image leaderboard",
     "url": "https://artificialanalysis.ai/image/leaderboard/text-to-image?v=20261011",
     "checked": "2026-10-11"
    },
    {
     "id": "s4",
     "title": "Artificial Analysis image editing leaderboard",
     "url": "https://artificialanalysis.ai/image/leaderboard/editing?v=20261011",
     "checked": "2026-10-11"
    },
    {
     "id": "s5",
     "title": "OpenAI API pricing",
     "url": "https://developers.openai.com/api/docs/pricing?v=20261011",
     "checked": "2026-10-11"
    },
    {
     "id": "s6",
     "title": "OpenAI: Introducing ChatGPT Images 2.5",
     "url": "https://openai.com/index/introducing-chatgpt-images-2-5/",
     "checked": "2026-10-11"
    },
    {
     "id": "s7",
     "title": "OpenAI enterprise privacy",
     "url": "https://openai.com/enterprise-privacy/",
     "checked": "2026-10-11"
    }
   ],
   "checked": "2026-10-11"
  },
  {
   "slug": "gpt-6-astra",
   "url": "https://hayzlabs.com/reviews/gpt-6-astra/",
   "name": "GPT-6 Astra",
   "maker": "OpenAI",
   "kind": "model",
   "license": "proprietary",
   "runs": "cloud",
   "released": "2026-09-03",
   "status": "graded",
   "verdict": "Depends",
   "verdict_line": "Second on WebDev and Agent boards, but GPT-6.1 Sol is close for a fifth of the price; middling at creative writing.",
   "grades": [
    {
     "category": "coding",
     "score": 78,
     "score_low": 65,
     "score_high": 92,
     "score_source": "model",
     "score_inputs": [
      "Arena",
      "Epoch AI"
     ],
     "usd_month": 20,
     "checked_on": "2026-10-11"
    },
    {
     "category": "agents",
     "score": null,
     "score_low": null,
     "score_high": null,
     "score_source": "model",
     "score_inputs": [
      "Arena"
     ],
     "arena_rank": 2,
     "usd_month": 20,
     "checked_on": "2026-10-11"
    },
    {
     "category": "research",
     "score": 93,
     "score_low": 88,
     "score_high": 96,
     "score_source": "model",
     "score_inputs": [
      "Arena",
      "Epoch AI"
     ],
     "usd_month": 20,
     "checked_on": "2026-10-11"
    },
    {
     "category": "writing",
     "score": 80,
     "score_low": 77,
     "score_high": 84,
     "score_source": "model",
     "score_inputs": [
      "Arena"
     ],
     "usd_month": 20,
     "checked_on": "2026-10-11"
    }
   ],
   "price": {
    "free_plan": false,
    "usd_month": 20,
    "plans": "ChatGPT Plus $20/month (s9) and Pro include GPT-6 Astra; Free and Go do not (s10).",
    "api": "$10 / $50 per 1M input / output tokens, cached input $1; long context $20 / $75 (s11)",
    "checked_on": "2026-10-11"
   },
   "how_we_rated": "From arena.ai leaderboards and the Artificial Analysis Intelligence Index, checked 11 Oct 2026. We did not run it ourselves.",
   "sources": [
    {
     "id": "s1",
     "title": "arena.ai Code Arena: WebDev leaderboard",
     "url": "https://arena.ai/leaderboard/code/webdev?v=20261011",
     "checked": "2026-10-11"
    },
    {
     "id": "s2",
     "title": "arena.ai Text: Coding leaderboard",
     "url": "https://arena.ai/leaderboard/chat/text/coding?v=20261011",
     "checked": "2026-10-11"
    },
    {
     "id": "s3",
     "title": "arena.ai Agent leaderboard",
     "url": "https://arena.ai/leaderboard/agent?v=20261011",
     "checked": "2026-10-11"
    },
    {
     "id": "s4",
     "title": "arena.ai Agent Pareto and cost per task",
     "url": "https://arena.ai/leaderboard/agent/pareto?v=20261011",
     "checked": "2026-10-11"
    },
    {
     "id": "s5",
     "title": "Artificial Analysis LLM leaderboard",
     "url": "https://artificialanalysis.ai/leaderboards/models?v=20261011",
     "checked": "2026-10-11"
    },
    {
     "id": "s6",
     "title": "arena.ai Text: Expert leaderboard",
     "url": "https://arena.ai/leaderboard/chat/text/expert?v=20261011",
     "checked": "2026-10-11"
    },
    {
     "id": "s7",
     "title": "arena.ai Text: Creative Writing leaderboard",
     "url": "https://arena.ai/leaderboard/chat/text/creative-writing?v=20261011",
     "checked": "2026-10-11"
    },
    {
     "id": "s8",
     "title": "arena.ai Text: Instruction Following leaderboard",
     "url": "https://arena.ai/leaderboard/chat/text/instruction-following?v=20261011",
     "checked": "2026-10-11"
    },
    {
     "id": "s9",
     "title": "OpenAI Help: What is ChatGPT Plus?",
     "url": "https://help.openai.com/en/articles/6950777-what-is-chatgpt-plus",
     "checked": "2026-10-11"
    },
    {
     "id": "s10",
     "title": "ChatGPT pricing and plan comparison",
     "url": "https://chatgpt.com/pricing",
     "checked": "2026-10-11"
    },
    {
     "id": "s11",
     "title": "OpenAI API pricing",
     "url": "https://developers.openai.com/api/docs/pricing",
     "checked": "2026-10-11"
    },
    {
     "id": "s12",
     "title": "OpenAI Help: How your data is used to improve model performance",
     "url": "https://help.openai.com/en/articles/5722486-how-your-data-is-used-to-improve-model-performance",
     "checked": "2026-10-11"
    },
    {
     "id": "s13",
     "title": "OpenAI API docs: Your data",
     "url": "https://developers.openai.com/api/docs/guides/your-data",
     "checked": "2026-10-11"
    },
    {
     "id": "s14",
     "title": "arena.ai Agent Alignment Index",
     "url": "https://arena.ai/leaderboard/agent/alignment?v=20261011",
     "checked": "2026-10-11"
    },
    {
     "id": "s15",
     "title": "OpenAI API docs: Models",
     "url": "https://developers.openai.com/api/docs/models",
     "checked": "2026-10-11"
    },
    {
     "id": "s16",
     "title": "OpenAI: Introducing GPT-6 Astra",
     "url": "https://openai.com/index/gpt-6-astra/",
     "checked": "2026-10-11"
    },
    {
     "id": "s17",
     "title": "Techmeme: OpenAI launches GPT-6 Astra",
     "url": "https://www.techmeme.com/260903/p46",
     "checked": "2026-10-11"
    }
   ],
   "checked": "2026-10-11"
  },
  {
   "slug": "mai-transcribe-2",
   "url": "https://hayzlabs.com/reviews/mai-transcribe-2/",
   "name": "MAI-Transcribe-2",
   "maker": "Microsoft",
   "kind": "model",
   "license": "proprietary",
   "runs": "cloud",
   "released": "2026-09-03",
   "status": "graded",
   "verdict": "Worth trying",
   "verdict_line": "Among the most accurate transcription models at $0.10 an hour. Still a preview with no SLA, so not for production yet.",
   "grades": [
    {
     "category": "voice",
     "score": null,
     "score_low": null,
     "score_high": null,
     "score_source": null,
     "usd_month": null,
     "checked_on": "2026-10-11"
    }
   ],
   "price": {
    "free_plan": false,
    "usd_month": null,
    "plans": "Pay as you go through Microsoft Foundry (Azure Speech); no monthly plan.",
    "api": "$0.10 per hour of audio, introductory price until the end of 2026; streaming version $0.54 per hour (s2, s3)",
    "checked_on": "2026-10-11"
   },
   "how_we_rated": "From Artificial Analysis speech to text word error rates, checked 11 Oct 2026. We did not run it ourselves.",
   "sources": [
    {
     "id": "s1",
     "title": "Artificial Analysis speech to text leaderboard",
     "url": "https://artificialanalysis.ai/speech-to-text/non-streaming",
     "checked": "2026-10-11"
    },
    {
     "id": "s2",
     "title": "Microsoft AI: MAI-Transcribe-2",
     "url": "https://microsoft.ai/models/mai-transcribe-2/",
     "checked": "2026-10-11"
    },
    {
     "id": "s3",
     "title": "Microsoft AI news: MAI-Transcribe-2 announcement",
     "url": "https://microsoft.ai/news/mai-transcribe-2-is-the-fastest-most-accurate-and-cheapest-speech-recognition-model-in-the-world/",
     "checked": "2026-10-11"
    },
    {
     "id": "s4",
     "title": "Microsoft Learn: MAI-Transcribe in Azure Speech (preview)",
     "url": "https://learn.microsoft.com/en-us/azure/ai-services/speech-service/mai-transcribe",
     "checked": "2026-10-11"
    },
    {
     "id": "s5",
     "title": "Microsoft Learn: speech to text data, privacy and security",
     "url": "https://learn.microsoft.com/en-us/azure/foundry/responsible-ai/speech-service/speech-to-text/data-privacy-security",
     "checked": "2026-10-11"
    }
   ],
   "checked": "2026-10-11"
  },
  {
   "slug": "gemini-3-8-flash",
   "url": "https://hayzlabs.com/reviews/gemini-3-8-flash/",
   "name": "Gemini 3.8 Flash",
   "maker": "Google",
   "kind": "model",
   "license": "proprietary",
   "runs": "cloud",
   "released": "2026-09-02",
   "status": "graded",
   "verdict": "Worth trying",
   "verdict_line": "Low-cost Google model near the top for writing and maths; rankings are preliminary and API prices double in January.",
   "grades": [
    {
     "category": "writing",
     "score": 91,
     "score_low": 89,
     "score_high": 93,
     "score_source": "model",
     "score_inputs": [
      "Arena"
     ],
     "usd_month": 19.99,
     "checked_on": "2026-10-11"
    },
    {
     "category": "data",
     "score": 85,
     "score_low": 80,
     "score_high": 88,
     "score_source": "model",
     "score_inputs": [
      "Arena",
      "Epoch AI"
     ],
     "usd_month": 19.99,
     "checked_on": "2026-10-11"
    },
    {
     "category": "research",
     "score": 88,
     "score_low": 84,
     "score_high": 92,
     "score_source": "model",
     "score_inputs": [
      "Arena",
      "Epoch AI"
     ],
     "usd_month": 19.99,
     "checked_on": "2026-10-11"
    },
    {
     "category": "coding",
     "score": 43,
     "score_low": 42,
     "score_high": 45,
     "score_source": "model",
     "score_inputs": [
      "Arena"
     ],
     "usd_month": 19.99,
     "checked_on": "2026-10-11"
    }
   ],
   "price": {
    "free_plan": false,
    "usd_month": 19.99,
    "plans": "In the Gemini app from Google AI Pro at $19.99/month; the Free and Plus plans list older Flash models (s8).",
    "api": "$0.75 / $3.75 per 1M input / output tokens through 31 Dec 2026, then $1.50 / $7.50 from 1 Jan 2027; free API tier available (s9)",
    "checked_on": "2026-10-11"
   },
   "how_we_rated": "From arena.ai leaderboards and the Artificial Analysis Intelligence Index, checked 11 Oct 2026. We did not run it ourselves.",
   "sources": [
    {
     "id": "s1",
     "title": "arena.ai Text: Creative Writing leaderboard",
     "url": "https://arena.ai/leaderboard/chat/text/creative-writing?v=20261011",
     "checked": "2026-10-11"
    },
    {
     "id": "s2",
     "title": "arena.ai Text: Instruction Following leaderboard",
     "url": "https://arena.ai/leaderboard/chat/text/instruction-following?v=20261011",
     "checked": "2026-10-11"
    },
    {
     "id": "s3",
     "title": "arena.ai Text: Math leaderboard",
     "url": "https://arena.ai/leaderboard/chat/text/math?v=20261011",
     "checked": "2026-10-11"
    },
    {
     "id": "s4",
     "title": "arena.ai Text: Expert leaderboard",
     "url": "https://arena.ai/leaderboard/chat/text/expert?v=20261011",
     "checked": "2026-10-11"
    },
    {
     "id": "s5",
     "title": "Artificial Analysis LLM leaderboard",
     "url": "https://artificialanalysis.ai/leaderboards/models?v=20261011",
     "checked": "2026-10-11"
    },
    {
     "id": "s6",
     "title": "arena.ai Code Arena: WebDev leaderboard",
     "url": "https://arena.ai/leaderboard/code/webdev?v=20261011",
     "checked": "2026-10-11"
    },
    {
     "id": "s7",
     "title": "arena.ai Text: Coding leaderboard",
     "url": "https://arena.ai/leaderboard/chat/text/coding?v=20261011",
     "checked": "2026-10-11"
    },
    {
     "id": "s8",
     "title": "Google AI plans (Gemini subscriptions)",
     "url": "https://gemini.google/subscriptions/",
     "checked": "2026-10-11"
    },
    {
     "id": "s9",
     "title": "Gemini API pricing",
     "url": "https://ai.google.dev/gemini-api/docs/pricing",
     "checked": "2026-10-11"
    },
    {
     "id": "s10",
     "title": "arena.ai Agent leaderboard",
     "url": "https://arena.ai/leaderboard/agent?v=20261011",
     "checked": "2026-10-11"
    },
    {
     "id": "s11",
     "title": "Gemini Apps Privacy Hub",
     "url": "https://support.google.com/gemini/answer/13594961",
     "checked": "2026-10-11"
    },
    {
     "id": "s12",
     "title": "Google DeepMind: Gemini 3.8 Flash model card",
     "url": "https://deepmind.google/models/model-cards/gemini-3-8-flash/",
     "checked": "2026-10-11"
    }
   ],
   "checked": "2026-10-11"
  },
  {
   "slug": "muse-spark-1-3",
   "url": "https://hayzlabs.com/reviews/muse-spark-1-3/",
   "name": "Muse Spark 1.3",
   "maker": "Meta",
   "kind": "model",
   "license": "proprietary",
   "runs": "cloud",
   "released": "2026-09-02",
   "status": "graded",
   "verdict": "Worth trying",
   "verdict_line": "Solid coding and agent model at a low token price; GPT-6.1 Sol and Claude rank higher if quality matters most.",
   "grades": [
    {
     "category": "coding",
     "score": 60,
     "score_low": 58,
     "score_high": 62,
     "score_source": "model",
     "score_inputs": [
      "Arena"
     ],
     "usd_month": 5,
     "checked_on": "2026-10-11"
    },
    {
     "category": "agents",
     "score": null,
     "score_low": null,
     "score_high": null,
     "score_source": "model",
     "score_inputs": [
      "Arena"
     ],
     "arena_rank": 15,
     "usd_month": 5,
     "checked_on": "2026-10-11"
    },
    {
     "category": "research",
     "score": 68,
     "score_low": 61,
     "score_high": 75,
     "score_source": "model",
     "score_inputs": [
      "Arena",
      "Epoch AI"
     ],
     "usd_month": 5,
     "checked_on": "2026-10-11"
    },
    {
     "category": "writing",
     "score": 83,
     "score_low": 81,
     "score_high": 86,
     "score_source": "model",
     "score_inputs": [
      "Arena"
     ],
     "usd_month": 5,
     "checked_on": "2026-10-11"
    }
   ],
   "price": {
    "free_plan": false,
    "usd_month": 5,
    "plans": "Muse Code plans from $5/month (Everyday Usage) with 'access to the latest Muse models' (s9); Axios reports that 1.3 rolled out in Muse Code and the API (s10).",
    "api": "$1.25 / $4.25 per 1M input / output tokens, cached input $0.15; a 'contributor' variant at $0.10 / $0.20 lets Meta use your data (s11)",
    "checked_on": "2026-10-11"
   },
   "how_we_rated": "From arena.ai leaderboards and the Artificial Analysis Intelligence Index, checked 11 Oct 2026. We did not run it ourselves.",
   "sources": [
    {
     "id": "s1",
     "title": "arena.ai Code Arena: WebDev leaderboard",
     "url": "https://arena.ai/leaderboard/code/webdev?v=20261011",
     "checked": "2026-10-11"
    },
    {
     "id": "s2",
     "title": "arena.ai Text: Coding leaderboard",
     "url": "https://arena.ai/leaderboard/chat/text/coding?v=20261011",
     "checked": "2026-10-11"
    },
    {
     "id": "s3",
     "title": "arena.ai Agent leaderboard",
     "url": "https://arena.ai/leaderboard/agent?v=20261011",
     "checked": "2026-10-11"
    },
    {
     "id": "s4",
     "title": "arena.ai Agent Pareto and cost per task",
     "url": "https://arena.ai/leaderboard/agent/pareto?v=20261011",
     "checked": "2026-10-11"
    },
    {
     "id": "s5",
     "title": "Artificial Analysis LLM leaderboard",
     "url": "https://artificialanalysis.ai/leaderboards/models?v=20261011",
     "checked": "2026-10-11"
    },
    {
     "id": "s6",
     "title": "arena.ai Text: Expert leaderboard",
     "url": "https://arena.ai/leaderboard/chat/text/expert?v=20261011",
     "checked": "2026-10-11"
    },
    {
     "id": "s7",
     "title": "arena.ai Text: Creative Writing leaderboard",
     "url": "https://arena.ai/leaderboard/chat/text/creative-writing?v=20261011",
     "checked": "2026-10-11"
    },
    {
     "id": "s8",
     "title": "arena.ai Text: Instruction Following leaderboard",
     "url": "https://arena.ai/leaderboard/chat/text/instruction-following?v=20261011",
     "checked": "2026-10-11"
    },
    {
     "id": "s9",
     "title": "Meta: Muse Code",
     "url": "https://dev.meta.ai/products/muse-code/",
     "checked": "2026-10-11"
    },
    {
     "id": "s10",
     "title": "Axios: Meta debuts Muse Spark 1.3",
     "url": "https://axios.com/2026/09/02/meta-debuts-muse-spark-13-as-personal-agent-work-continues",
     "checked": "2026-10-11"
    },
    {
     "id": "s11",
     "title": "Meta Model API: Muse Spark",
     "url": "https://dev.meta.ai/models/muse-spark",
     "checked": "2026-10-11"
    },
    {
     "id": "s12",
     "title": "arena.ai Agent Alignment Index",
     "url": "https://arena.ai/leaderboard/agent/alignment?v=20261011",
     "checked": "2026-10-11"
    }
   ],
   "checked": "2026-10-11"
  },
  {
   "slug": "claude-fable-5-1",
   "url": "https://hayzlabs.com/reviews/claude-fable-5-1/",
   "name": "Claude Fable 5.1",
   "maker": "Anthropic",
   "kind": "model",
   "license": "proprietary",
   "runs": "cloud",
   "released": "2026-09-01",
   "status": "graded",
   "verdict": "Skip",
   "verdict_line": "Very strong, but Opus 5.5 ranks higher on most boards for 40% of the price; keep Fable for cases Opus misses.",
   "grades": [
    {
     "category": "agents",
     "score": null,
     "score_low": null,
     "score_high": null,
     "score_source": "model",
     "score_inputs": [
      "Arena"
     ],
     "arena_rank": 3,
     "usd_month": 100,
     "checked_on": "2026-10-11"
    },
    {
     "category": "coding",
     "score": 87,
     "score_low": 76,
     "score_high": 90,
     "score_source": "model",
     "score_inputs": [
      "Arena",
      "Epoch AI"
     ],
     "usd_month": 100,
     "checked_on": "2026-10-11"
    },
    {
     "category": "research",
     "score": 88,
     "score_low": 82,
     "score_high": 94,
     "score_source": "model",
     "score_inputs": [
      "Arena",
      "Epoch AI"
     ],
     "usd_month": 100,
     "checked_on": "2026-10-11"
    }
   ],
   "price": {
    "free_plan": false,
    "usd_month": 100,
    "plans": "Max from $100/month includes Fable (listed as 50% of weekly limits); on Pro it runs on usage credits; not on Free (s7).",
    "api": "$10 / $50 per 1M input / output tokens (s8)",
    "checked_on": "2026-10-11"
   },
   "how_we_rated": "From arena.ai leaderboards and the Artificial Analysis Intelligence Index, checked 11 Oct 2026. We did not run it ourselves.",
   "sources": [
    {
     "id": "s1",
     "title": "arena.ai Agent leaderboard",
     "url": "https://arena.ai/leaderboard/agent?v=20261011",
     "checked": "2026-10-11"
    },
    {
     "id": "s2",
     "title": "arena.ai Agent Pareto and cost per task",
     "url": "https://arena.ai/leaderboard/agent/pareto?v=20261011",
     "checked": "2026-10-11"
    },
    {
     "id": "s3",
     "title": "Artificial Analysis LLM leaderboard",
     "url": "https://artificialanalysis.ai/leaderboards/models?v=20261011",
     "checked": "2026-10-11"
    },
    {
     "id": "s4",
     "title": "arena.ai Code Arena: WebDev leaderboard",
     "url": "https://arena.ai/leaderboard/code/webdev?v=20261011",
     "checked": "2026-10-11"
    },
    {
     "id": "s5",
     "title": "arena.ai Text: Coding leaderboard",
     "url": "https://arena.ai/leaderboard/chat/text/coding?v=20261011",
     "checked": "2026-10-11"
    },
    {
     "id": "s6",
     "title": "arena.ai Text: Expert leaderboard",
     "url": "https://arena.ai/leaderboard/chat/text/expert?v=20261011",
     "checked": "2026-10-11"
    },
    {
     "id": "s7",
     "title": "Claude plans and pricing",
     "url": "https://claude.com/pricing",
     "checked": "2026-10-11"
    },
    {
     "id": "s8",
     "title": "Claude API pricing",
     "url": "https://platform.claude.com/docs/en/about-claude/pricing",
     "checked": "2026-10-11"
    },
    {
     "id": "s9",
     "title": "Claude privacy: is my data used for model training (consumer)",
     "url": "https://privacy.claude.com/en/articles/10023580-is-my-data-used-for-model-training",
     "checked": "2026-10-11"
    },
    {
     "id": "s10",
     "title": "Claude privacy: is my data used for model training (commercial)",
     "url": "https://privacy.claude.com/en/articles/7996868-is-my-data-used-for-model-training",
     "checked": "2026-10-11"
    },
    {
     "id": "s11",
     "title": "arena.ai Text: Creative Writing leaderboard",
     "url": "https://arena.ai/leaderboard/chat/text/creative-writing?v=20261011",
     "checked": "2026-10-11"
    },
    {
     "id": "s12",
     "title": "Claude Fable 5.1 model overview",
     "url": "https://platform.claude.com/docs/en/models/fable-5-1/overview",
     "checked": "2026-10-11"
    },
    {
     "id": "s13",
     "title": "Claude models overview",
     "url": "https://platform.claude.com/docs/en/models/overview",
     "checked": "2026-10-11"
    }
   ],
   "checked": "2026-10-11"
  },
  {
   "slug": "claude-mythos-5-1",
   "url": "https://hayzlabs.com/reviews/claude-mythos-5-1/",
   "name": "Claude Mythos 5.1",
   "maker": "Anthropic",
   "kind": "model",
   "license": "proprietary",
   "runs": "cloud",
   "released": "2026-09-01",
   "status": "first-look",
   "verdict": "Wait",
   "verdict_line": "Anthropic's top cyber and biology model, open only to vetted groups. Most people should use Opus 5.5.",
   "grades": [],
   "price": {
    "free_plan": false,
    "usd_month": null,
    "plans": "No consumer plan. Access only through Anthropic's trusted access programs (s1, s2).",
    "api": "$10 / $50 per 1M input / output tokens; cache hits $0.25; Batch $5 / $25 (s3)",
    "checked_on": "2026-10-11"
   },
   "how_we_rated": "First look from Anthropic's pages, checked 11 Oct 2026. Not rated yet. We did not run it ourselves.",
   "sources": [
    {
     "id": "s1",
     "title": "Anthropic: Claude Mythos",
     "url": "https://www.anthropic.com/claude/mythos?v=20261011",
     "checked": "2026-10-11"
    },
    {
     "id": "s2",
     "title": "Anthropic: Project Glasswing",
     "url": "https://www.anthropic.com/glasswing?v=20261011",
     "checked": "2026-10-11"
    },
    {
     "id": "s3",
     "title": "Claude API pricing",
     "url": "https://platform.claude.com/docs/en/about-claude/pricing?v=20261011",
     "checked": "2026-10-11"
    },
    {
     "id": "s4",
     "title": "arena.ai leaderboards",
     "url": "https://arena.ai/leaderboard?v=20261011",
     "checked": "2026-10-11"
    },
    {
     "id": "s5",
     "title": "Artificial Analysis LLM leaderboard",
     "url": "https://artificialanalysis.ai/leaderboards/models?v=20261011",
     "checked": "2026-10-11"
    },
    {
     "id": "s6",
     "title": "Claude model deprecations (model status table)",
     "url": "https://platform.claude.com/docs/en/about-claude/model-deprecations?v=20261011",
     "checked": "2026-10-11"
    }
   ],
   "checked": "2026-10-11"
  },
  {
   "slug": "gemini-omni-flash-1-1",
   "url": "https://hayzlabs.com/reviews/gemini-omni-flash-1-1/",
   "name": "Gemini Omni Flash 1.1",
   "maker": "Google",
   "kind": "model",
   "license": "proprietary",
   "runs": "cloud",
   "released": "2026-08-27",
   "status": "graded",
   "verdict": "Use it",
   "verdict_line": "Top-rated text to video on arena.ai, from $4.99/month in Google Flow. Skip it if price per clip matters most.",
   "grades": [
    {
     "category": "video",
     "score": 100,
     "score_low": 96,
     "score_high": 100,
     "score_source": "model",
     "score_inputs": [
      "Arena"
     ],
     "usd_month": 4.99,
     "checked_on": "2026-10-11"
    }
   ],
   "price": {
    "free_plan": false,
    "usd_month": 4.99,
    "plans": "Google AI Plus $4.99/month includes 200 Google Flow credits with access to Gemini Omni Flash; Pro $19.99/month includes 1,000 (s5)",
    "api": "$1.50 per 1M input tokens; video output $17.50 per 1M tokens, about $0.10 per second of 720p; no free API tier (s6)",
    "checked_on": "2026-10-11"
   },
   "how_we_rated": "From the arena.ai video leaderboards and the Artificial Analysis video arena, checked 11 Oct 2026. We did not run it ourselves.",
   "sources": [
    {
     "id": "s1",
     "title": "arena.ai Text to Video leaderboard",
     "url": "https://arena.ai/leaderboard/video/text-to-video?v=20261011",
     "checked": "2026-10-11"
    },
    {
     "id": "s2",
     "title": "arena.ai Image to Video leaderboard",
     "url": "https://arena.ai/leaderboard/video/image-to-video?v=20261011",
     "checked": "2026-10-11"
    },
    {
     "id": "s3",
     "title": "Artificial Analysis text to video leaderboard",
     "url": "https://artificialanalysis.ai/video/leaderboard/text-to-video?v=20261011",
     "checked": "2026-10-11"
    },
    {
     "id": "s4",
     "title": "Artificial Analysis image to video leaderboard",
     "url": "https://artificialanalysis.ai/video/leaderboard/image-to-video?v=20261011",
     "checked": "2026-10-11"
    },
    {
     "id": "s5",
     "title": "Google AI plans (Gemini subscriptions)",
     "url": "https://gemini.google/subscriptions/?v=20261011",
     "checked": "2026-10-11"
    },
    {
     "id": "s6",
     "title": "Gemini API pricing",
     "url": "https://ai.google.dev/gemini-api/docs/pricing?v=20261011",
     "checked": "2026-10-11"
    },
    {
     "id": "s7",
     "title": "Gemini Apps Privacy Hub",
     "url": "https://support.google.com/gemini/answer/13594961?hl=en",
     "checked": "2026-10-11"
    },
    {
     "id": "s8",
     "title": "Google blog: Build with Gemini Omni 1.1 Flash",
     "url": "https://blog.google/innovation-and-ai/technology/developers-tools/build-with-gemini-omni-1-1-flash/",
     "checked": "2026-10-11"
    },
    {
     "id": "s9",
     "title": "Google blog: Gemini Omni",
     "url": "https://blog.google/innovation-and-ai/models-and-research/gemini-models/gemini-omni/",
     "checked": "2026-10-11"
    }
   ],
   "checked": "2026-10-11"
  },
  {
   "slug": "glm-5-3-flash",
   "url": "https://hayzlabs.com/reviews/glm-5-3-flash/",
   "name": "GLM-5.3-Flash",
   "maker": "Z.ai (Zhipu)",
   "kind": "open-source",
   "license": "open-source",
   "runs": "cloud",
   "released": "2026-08-25",
   "status": "graded",
   "verdict": "Worth trying",
   "verdict_line": "Very cheap MIT model, 13th on arena.ai Math; for coding and agents MiMo-V2.6-Pro ranks higher for less per task.",
   "grades": [
    {
     "category": "data",
     "score": 79,
     "score_low": 73,
     "score_high": 84,
     "score_source": "model",
     "score_inputs": [
      "Arena",
      "Epoch AI"
     ],
     "usd_month": 0,
     "checked_on": "2026-10-11"
    },
    {
     "category": "research",
     "score": 74,
     "score_low": 70,
     "score_high": 77,
     "score_source": "model",
     "score_inputs": [
      "Arena",
      "Epoch AI"
     ],
     "usd_month": 0,
     "checked_on": "2026-10-11"
    },
    {
     "category": "coding",
     "score": 48,
     "score_low": 47,
     "score_high": 49,
     "score_source": "model",
     "score_inputs": [
      "Arena"
     ],
     "usd_month": 0,
     "checked_on": "2026-10-11"
    }
   ],
   "price": {
    "free_plan": false,
    "usd_month": 0,
    "plans": "Chat with it in Z.ai's app at chat.z.ai; the app does not state a price (s10). Free weights under MIT (s6). Z.ai API: $0.15 input, $0.03 cached input, $0.50 output per 1M tokens (s7).",
    "api": "$0.15 / $0.50 per 1M input / output tokens on Z.ai (s7)",
    "checked_on": "2026-10-11"
   },
   "how_we_rated": "From arena.ai leaderboards and the Artificial Analysis Intelligence Index, checked 11 Oct 2026. We did not run it ourselves.",
   "sources": [
    {
     "id": "s10",
     "title": "Z.ai chat app",
     "url": "https://chat.z.ai/",
     "checked": "2026-10-11"
    },
    {
     "id": "s1",
     "title": "arena.ai Text: Math leaderboard",
     "url": "https://arena.ai/leaderboard/chat/text/math?v=20261011",
     "checked": "2026-10-11"
    },
    {
     "id": "s2",
     "title": "arena.ai Text: Expert leaderboard",
     "url": "https://arena.ai/leaderboard/chat/text/expert?v=20261011",
     "checked": "2026-10-11"
    },
    {
     "id": "s3",
     "title": "Artificial Analysis LLM leaderboard",
     "url": "https://artificialanalysis.ai/leaderboards/models?v=20261011",
     "checked": "2026-10-11"
    },
    {
     "id": "s4",
     "title": "arena.ai Code Arena: WebDev leaderboard",
     "url": "https://arena.ai/leaderboard/code/webdev?v=20261011",
     "checked": "2026-10-11"
    },
    {
     "id": "s5",
     "title": "arena.ai Text: Coding leaderboard",
     "url": "https://arena.ai/leaderboard/chat/text/coding?v=20261011",
     "checked": "2026-10-11"
    },
    {
     "id": "s6",
     "title": "Hugging Face: zai-org/GLM-5.3-Flash model card",
     "url": "https://huggingface.co/zai-org/GLM-5.3-Flash",
     "checked": "2026-10-11"
    },
    {
     "id": "s7",
     "title": "Z.ai pricing",
     "url": "https://docs.z.ai/guides/overview/pricing",
     "checked": "2026-10-11"
    },
    {
     "id": "s8",
     "title": "Z.ai privacy policy and API data processing terms",
     "url": "https://docs.z.ai/legal-agreement/privacy-policy",
     "checked": "2026-10-11"
    },
    {
     "id": "s9",
     "title": "arena.ai Text: Creative Writing leaderboard",
     "url": "https://arena.ai/leaderboard/chat/text/creative-writing?v=20261011",
     "checked": "2026-10-11"
    }
   ],
   "checked": "2026-10-11"
  },
  {
   "slug": "wan-3-0",
   "url": "https://hayzlabs.com/reviews/wan-3-0/",
   "name": "Wan 3.0",
   "maker": "Alibaba",
   "kind": "model",
   "license": "proprietary",
   "runs": "cloud",
   "released": "2026-08-24",
   "status": "graded",
   "verdict": "Use it",
   "verdict_line": "Top-tier video from text or images, clips up to 30 seconds, at half Veo 3.1's price. MiniMax H3 costs less.",
   "grades": [
    {
     "category": "video",
     "score": 89,
     "score_low": 85,
     "score_high": 92,
     "score_source": "model",
     "score_inputs": [
      "Arena"
     ],
     "usd_month": null,
     "checked_on": "2026-10-11"
    }
   ],
   "price": {
    "free_plan": false,
    "usd_month": null,
    "plans": "Pay as you go on Alibaba Cloud Model Studio; no monthly plan. About $9.90 per minute of 1080p in the US (Virginia) region, $12 in Singapore.",
    "api": "$0.20 / $0.10 / $0.05 per second at 1080p / 720p / 480p (Singapore list price); $0.165 / $0.083 / $0.041 per second in US, Germany, Japan and other regions (s5)",
    "checked_on": "2026-10-11"
   },
   "how_we_rated": "From the arena.ai video leaderboards and the Artificial Analysis video arena, checked 11 Oct 2026. We did not run it ourselves.",
   "sources": [
    {
     "id": "s1",
     "title": "arena.ai Image to Video leaderboard",
     "url": "https://arena.ai/leaderboard/video/image-to-video",
     "checked": "2026-10-11"
    },
    {
     "id": "s2",
     "title": "arena.ai Text to Video leaderboard",
     "url": "https://arena.ai/leaderboard/video/text-to-video",
     "checked": "2026-10-11"
    },
    {
     "id": "s3",
     "title": "Artificial Analysis text to video leaderboard",
     "url": "https://artificialanalysis.ai/video/leaderboard/text-to-video",
     "checked": "2026-10-11"
    },
    {
     "id": "s4",
     "title": "Artificial Analysis image to video leaderboard",
     "url": "https://artificialanalysis.ai/video/leaderboard/image-to-video",
     "checked": "2026-10-11"
    },
    {
     "id": "s5",
     "title": "Alibaba Cloud Model Studio: Wan3.0 video",
     "url": "https://www.alibabacloud.com/help/en/model-studio/wan3-0-video",
     "checked": "2026-10-11"
    },
    {
     "id": "s6",
     "title": "Alibaba Cloud Model Studio privacy notice",
     "url": "https://www.alibabacloud.com/help/en/model-studio/privacy-notice",
     "checked": "2026-10-11"
    },
    {
     "id": "s7",
     "title": "The Next Web: Alibaba launches Wan3.0 video model",
     "url": "https://thenextweb.com/news/alibaba-wan3-video-model-after-share-sale",
     "checked": "2026-10-11"
    }
   ],
   "checked": "2026-10-11"
  },
  {
   "slug": "microsoft-copilot",
   "url": "https://hayzlabs.com/reviews/microsoft-copilot/",
   "name": "Microsoft Copilot",
   "maker": "Microsoft",
   "kind": "app",
   "license": "proprietary",
   "runs": "cloud",
   "released": "2026-08-18",
   "status": "first-look",
   "verdict": "Worth trying",
   "verdict_line": "Free Microsoft assistant; 365 plans add it to Word and Excel. No score yet: Microsoft doesn't name its models.",
   "grades": [],
   "price": {
    "free_plan": true,
    "usd_month": 0,
    "plans": "Free Copilot app, usage limits not published. Microsoft 365 Personal $9.99/month or $99.99/year: 'Advanced Copilot chat with higher usage than free', plus Copilot in Word, Excel, PowerPoint and Outlook (the pricing page's app list marks Word, Excel and PowerPoint 'web and mobile only' on this plan, while its feature list says desktop apps). Family $12.99/month or $129.99/year, with Copilot in those apps for the subscription owner only. Premium $19.99/month or $199.99/year: 'extensive usage' and 'Limited usage for agentic AI' for research and analytics. Pro $99.99/month, monthly billing only: 'highest usage' (s1, s2).",
    "api": null,
    "checked_on": "2026-10-18"
   },
   "how_we_rated": "First look from Microsoft's plan pages and its Copilot privacy pages, checked 18 Oct 2026. Not rated yet. We did not run it ourselves.",
   "sources": [
    {
     "id": "s1",
     "title": "Microsoft Copilot plans for individuals",
     "url": "https://www.microsoft.com/en-us/copilot/pricing/individuals?v=1018e",
     "checked": "2026-10-18"
    },
    {
     "id": "s2",
     "title": "Microsoft 365 Personal, Family and Premium compared",
     "url": "https://www.microsoft.com/en-us/microsoft-365/premium?v=1018e",
     "checked": "2026-10-18"
    },
    {
     "id": "s3",
     "title": "Privacy FAQ for Microsoft Copilot (older app)",
     "url": "https://support.microsoft.com/en-us/topic/privacy-faq-for-microsoft-copilot-27b3a435-8dc9-4b55-9a4b-58eeb9647a7f?v=1018e",
     "checked": "2026-10-18"
    },
    {
     "id": "s4",
     "title": "Microsoft Copilot for individuals: your data, privacy, and responsible AI",
     "url": "https://support.microsoft.com/en-us/privacy/microsoft-copilot/overview?v=1018e",
     "checked": "2026-10-18"
    },
    {
     "id": "s5",
     "title": "Microsoft Copilot privacy controls (new app)",
     "url": "https://support.microsoft.com/en-us/privacy/microsoft-copilot/privacy-controls?v=1018e",
     "checked": "2026-10-18"
    },
    {
     "id": "s6",
     "title": "Microsoft Copilot transparency note",
     "url": "https://support.microsoft.com/en-us/privacy/microsoft-copilot/transparency-note?v=1018e",
     "checked": "2026-10-18"
    }
   ],
   "checked": "2026-10-18"
  },
  {
   "slug": "minimax-h3",
   "url": "https://hayzlabs.com/reviews/minimax-h3/",
   "name": "MiniMax H3",
   "maker": "MiniMax",
   "kind": "model",
   "license": "open-weights",
   "runs": "both",
   "released": "2026-07-31",
   "status": "graded",
   "verdict": "Use it",
   "verdict_line": "Best value in video: top-rated image to video with sound at about $5 a minute. Weights barred in US, EU, UK.",
   "grades": [
    {
     "category": "video",
     "score": 84,
     "score_low": 82,
     "score_high": 86,
     "score_source": "model",
     "score_inputs": [
      "Arena"
     ],
     "usd_month": null,
     "checked_on": "2026-10-11"
    }
   ],
   "price": {
    "free_plan": false,
    "usd_month": null,
    "plans": "Pay as you go on the MiniMax API; no monthly plan. About $4.80 per minute at 768p.",
    "api": "MiniMax-H3: $0.08 per second at 768p, $0.13 per second at 2K. MiniMax-H3-Max: $0.08 per second at 768p, $0.05 at 480p. Extra input images are billed separately (s5).",
    "checked_on": "2026-10-11"
   },
   "how_we_rated": "From the arena.ai video leaderboards and the Artificial Analysis video arena, checked 11 Oct 2026. We did not run it ourselves.",
   "sources": [
    {
     "id": "s1",
     "title": "arena.ai Image to Video leaderboard",
     "url": "https://arena.ai/leaderboard/video/image-to-video",
     "checked": "2026-10-11"
    },
    {
     "id": "s2",
     "title": "arena.ai Text to Video leaderboard",
     "url": "https://arena.ai/leaderboard/video/text-to-video",
     "checked": "2026-10-11"
    },
    {
     "id": "s3",
     "title": "Artificial Analysis image to video leaderboard",
     "url": "https://artificialanalysis.ai/video/leaderboard/image-to-video",
     "checked": "2026-10-11"
    },
    {
     "id": "s4",
     "title": "Artificial Analysis text to video leaderboard",
     "url": "https://artificialanalysis.ai/video/leaderboard/text-to-video",
     "checked": "2026-10-11"
    },
    {
     "id": "s5",
     "title": "MiniMax API pay as you go pricing",
     "url": "https://platform.minimax.io/docs/guides/pricing-paygo.md",
     "checked": "2026-10-11"
    },
    {
     "id": "s6",
     "title": "MiniMax blog: MiniMax H3",
     "url": "https://www.minimax.io/blog/minimax-h3",
     "checked": "2026-10-11"
    },
    {
     "id": "s7",
     "title": "MiniMax docs: run and self-host MiniMax H3",
     "url": "https://platform.minimax.io/docs/guides/local-deploy-h3.md",
     "checked": "2026-10-11"
    }
   ],
   "checked": "2026-10-11"
  },
  {
   "slug": "seedance-2-5",
   "url": "https://hayzlabs.com/reviews/seedance-2-5/",
   "name": "Seedance 2.5",
   "maker": "ByteDance",
   "kind": "model",
   "license": "proprietary",
   "runs": "cloud",
   "released": "2026-07-31",
   "status": "graded",
   "verdict": "Worth trying",
   "verdict_line": "Top-tier video with 30 second clips, but about 7x MiniMax H3's price per minute. Most people should start with H3.",
   "grades": [
    {
     "category": "video",
     "score": 88,
     "score_low": 85,
     "score_high": 90,
     "score_source": "model",
     "score_inputs": [
      "Arena"
     ],
     "usd_month": null,
     "checked_on": "2026-10-11"
    }
   ],
   "price": {
    "free_plan": false,
    "usd_month": null,
    "plans": "Offered through ByteDance's Dola app and the BytePlus API (s6); we could not confirm plan or API prices on ByteDance or BytePlus pages.",
    "api": null,
    "checked_on": "2026-10-11"
   },
   "how_we_rated": "From the arena.ai video leaderboards and the Artificial Analysis video arena, checked 11 Oct 2026. We did not run it ourselves.",
   "sources": [
    {
     "id": "s1",
     "title": "arena.ai Text to Video leaderboard",
     "url": "https://arena.ai/leaderboard/video/text-to-video?v=20261011",
     "checked": "2026-10-11"
    },
    {
     "id": "s2",
     "title": "arena.ai Image to Video leaderboard",
     "url": "https://arena.ai/leaderboard/video/image-to-video?v=20261011",
     "checked": "2026-10-11"
    },
    {
     "id": "s3",
     "title": "Artificial Analysis text to video leaderboard",
     "url": "https://artificialanalysis.ai/video/leaderboard/text-to-video?v=20261011",
     "checked": "2026-10-11"
    },
    {
     "id": "s4",
     "title": "Artificial Analysis image to video leaderboard",
     "url": "https://artificialanalysis.ai/video/leaderboard/image-to-video?v=20261011",
     "checked": "2026-10-11"
    },
    {
     "id": "s5",
     "title": "arena.ai Video Edit leaderboard",
     "url": "https://arena.ai/leaderboard/video/video-edit",
     "checked": "2026-10-11"
    },
    {
     "id": "s6",
     "title": "ByteDance Seed: Seedance 2.5",
     "url": "https://seed.bytedance.com/en/seedance2_5",
     "checked": "2026-10-11"
    },
    {
     "id": "s7",
     "title": "TechNode: ByteDance launches Seedance 2.5 video generation model",
     "url": "https://technode.com/2026/07/31/bytedance-launches-seedance-2-5-video-generation-model/",
     "checked": "2026-10-11"
    }
   ],
   "checked": "2026-10-11"
  },
  {
   "slug": "meta-ai",
   "url": "https://hayzlabs.com/reviews/meta-ai/",
   "name": "Meta AI",
   "maker": "Meta",
   "kind": "app",
   "license": "proprietary",
   "runs": "cloud",
   "released": "2026-07-24",
   "status": "first-look",
   "verdict": "Wait",
   "verdict_line": "Free assistant in Meta's apps; your chats can shape your ads. No score yet: Meta doesn't say which model runs now.",
   "grades": [],
   "price": {
    "free_plan": true,
    "usd_month": 0,
    "plans": "Free: 'Meta AI will still be free for everyday use'. Meta One Core $7.99/month: 'generate more images and videos with Meta AI'. Meta One Premium $19.99/month: 'the most room to create content with Meta AI'. Meta says plans, prices and availability 'may vary by region, by app, and by account' (s1).",
    "api": null,
    "checked_on": "2026-10-18"
   },
   "how_we_rated": "First look from Meta's newsroom posts on Meta One, Meta AI, Muse and ads, checked 18 Oct 2026. Meta's help pages on AI data could not be read. Not rated yet. We did not run it ourselves.",
   "sources": [
    {
     "id": "s1",
     "title": "Meta: Introducing Meta One",
     "url": "https://about.fb.com/news/2026/09/introducing-meta-one-subscription-service-more-features-ai/?v=1018e",
     "checked": "2026-10-18"
    },
    {
     "id": "s2",
     "title": "Meta: Improving your recommendations on our apps with AI at Meta",
     "url": "https://about.fb.com/news/2025/10/improving-your-recommendations-apps-ai-meta/?v=1018e",
     "checked": "2026-10-18"
    },
    {
     "id": "s3",
     "title": "Meta: Meta AI doesn't just think, it acts",
     "url": "https://about.fb.com/news/2026/07/meta-ai-muse-spark-doesnt-just-think-it-acts/?v=1018e",
     "checked": "2026-10-18"
    },
    {
     "id": "s4",
     "title": "Meta: Introducing Muse, a personal AI agent",
     "url": "https://about.fb.com/news/2026/09/introducing-muse-personal-ai-agent/?v=1018e",
     "checked": "2026-10-18"
    },
    {
     "id": "s5",
     "title": "Meta: Introducing Muse Wonder and Muse Wonder Full",
     "url": "https://about.fb.com/news/2026/10/introducing-muse-wonder-models/?v=1018e",
     "checked": "2026-10-18"
    },
    {
     "id": "s6",
     "title": "Meta Help Center: Manage your information on Meta AI and Vibes (no readable text)",
     "url": "https://www.meta.com/help/artificial-intelligence/1771195753735844/?v=1018e",
     "checked": "2026-10-18"
    }
   ],
   "checked": "2026-10-18"
  },
  {
   "slug": "midjourney-v8-2",
   "url": "https://hayzlabs.com/reviews/midjourney-v8-2/",
   "name": "Midjourney V8.2",
   "maker": "Midjourney",
   "kind": "model",
   "license": "proprietary",
   "runs": "cloud",
   "released": "2026-07-24",
   "status": "first-look",
   "verdict": "Worth trying",
   "verdict_line": "Paid image tool for a distinctive look; no blind leaderboard ranks it, and images are public below the $60 Pro plan.",
   "grades": [],
   "price": {
    "free_plan": false,
    "usd_month": 10,
    "plans": "Basic $10/month, Standard $30, Pro $60, Mega $120; 20% off yearly. Unlimited Relax mode from Standard; Stealth (private) mode only on Pro and Mega (s2)",
    "api": "None",
    "checked_on": "2026-10-11"
   },
   "how_we_rated": "First look from Midjourney's own docs, plans and terms, checked 11 Oct 2026. Not rated yet: no independent leaderboard lists it. We did not run it ourselves.",
   "sources": [
    {
     "id": "s1",
     "title": "Midjourney docs: Version (models)",
     "url": "https://docs.midjourney.com/docs/models",
     "checked": "2026-10-11"
    },
    {
     "id": "s2",
     "title": "Midjourney docs: Comparing Midjourney Plans",
     "url": "https://docs.midjourney.com/hc/en-us/articles/27870484040333-Comparing-Midjourney-Plans",
     "checked": "2026-10-11"
    },
    {
     "id": "s3",
     "title": "Midjourney Terms of Service",
     "url": "https://docs.midjourney.com/hc/en-us/articles/32083055291277-Terms-of-Service",
     "checked": "2026-10-11"
    },
    {
     "id": "s4",
     "title": "arena.ai Text to Image leaderboard",
     "url": "https://arena.ai/leaderboard/image/text-to-image?v=20261011",
     "checked": "2026-10-11"
    },
    {
     "id": "s5",
     "title": "Artificial Analysis text-to-image leaderboard",
     "url": "https://artificialanalysis.ai/image/leaderboard/text-to-image?v=20261011",
     "checked": "2026-10-11"
    }
   ],
   "checked": "2026-10-11"
  },
  {
   "slug": "flux-3-video",
   "url": "https://hayzlabs.com/reviews/flux-3-video/",
   "name": "FLUX 3 Video",
   "maker": "Black Forest Labs",
   "kind": "model",
   "license": "proprietary",
   "runs": "cloud",
   "released": "2026-07-23",
   "status": "graded",
   "verdict": "Worth trying",
   "verdict_line": "Strong new video model with sound and cheap drafts, but pricier than equals. MiniMax H3 gives more per dollar.",
   "grades": [
    {
     "category": "video",
     "score": 93,
     "score_low": 88,
     "score_high": 98,
     "score_source": "model",
     "score_inputs": [
      "Arena"
     ],
     "usd_month": null,
     "checked_on": "2026-10-11"
    }
   ],
   "price": {
    "free_plan": false,
    "usd_month": null,
    "plans": "Pay as you go on the BFL dashboard and API; no monthly plan. About $17.40 per minute at 1080p (fhd) (s3, s4).",
    "api": "Text or image to video: $0.17 / $0.29 / $0.40 / $0.80 per second at hd / fhd / qhd / uhd; drafts $0.06 per second (s4)",
    "checked_on": "2026-10-11"
   },
   "how_we_rated": "From the arena.ai video leaderboards and the Artificial Analysis video arena, checked 11 Oct 2026. We did not run it ourselves.",
   "sources": [
    {
     "id": "s1",
     "title": "arena.ai Text to Video leaderboard",
     "url": "https://arena.ai/leaderboard/video/text-to-video",
     "checked": "2026-10-11"
    },
    {
     "id": "s2",
     "title": "arena.ai Image to Video leaderboard",
     "url": "https://arena.ai/leaderboard/video/image-to-video",
     "checked": "2026-10-11"
    },
    {
     "id": "s3",
     "title": "Artificial Analysis text to video leaderboard",
     "url": "https://artificialanalysis.ai/video/leaderboard/text-to-video",
     "checked": "2026-10-11"
    },
    {
     "id": "s4",
     "title": "Black Forest Labs API pricing",
     "url": "https://docs.bfl.ai/quick_start/pricing",
     "checked": "2026-10-11"
    },
    {
     "id": "s5",
     "title": "Black Forest Labs terms of service",
     "url": "https://bfl.ai/legal/terms-of-service",
     "checked": "2026-10-11"
    },
    {
     "id": "s6",
     "title": "Black Forest Labs: FLUX 3 model page",
     "url": "https://bfl.ai/models/flux-3-video-video",
     "checked": "2026-10-11"
    },
    {
     "id": "s7",
     "title": "Black Forest Labs blog: FLUX 3",
     "url": "https://bfl.ai/blog/flux-3",
     "checked": "2026-10-11"
    }
   ],
   "checked": "2026-10-11"
  },
  {
   "slug": "lm-studio-with-bionic-agent",
   "url": "https://hayzlabs.com/reviews/lm-studio-with-bionic-agent/",
   "name": "LM Studio (with Bionic agent)",
   "maker": "LM Studio",
   "kind": "app",
   "license": "proprietary",
   "runs": "both",
   "released": "2026-07-17",
   "status": "graded",
   "verdict": "Worth trying",
   "verdict_line": "Free desktop app for running open coding models privately, now with a local agent; skip it if you need frontier quality.",
   "grades": [
    {
     "category": "coding",
     "score": 80,
     "score_low": 80,
     "score_high": 80,
     "score_source": "checklist",
     "usd_month": 0,
     "checked_on": "2026-10-18"
    }
   ],
   "price": {
    "free_plan": true,
    "usd_month": 0,
    "plans": "Free (local models, Bionic agent, limited web search); Bionic+ $20/month adds hosted open models; Pro $100/month (s4)",
    "api": null,
    "checked_on": "2026-10-11"
   },
   "how_we_rated": "From the SPEC checklist, using an independent speed measurement, arena.ai ranks of the models it can run, and LM Studio's own pricing, privacy and changelog pages, checked 11 Oct 2026. We did not run it ourselves.",
   "sources": [
    {
     "id": "s1",
     "title": "Mozilla.ai: Benchmarking llama.cpp vs llamafile vs LM Studio vs Ollama, 17 Sep 2026",
     "url": "https://blog.mozilla.ai/benchmarking-local-llm-servers-llama-cpp-llamafile-lm-studio-and-ollama/",
     "checked": "2026-10-18"
    },
    {
     "id": "s2",
     "title": "LM Studio model catalog",
     "url": "https://lmstudio.ai/models?v=20261011",
     "checked": "2026-10-11"
    },
    {
     "id": "s3",
     "title": "arena.ai Code Arena: WebDev leaderboard",
     "url": "https://arena.ai/leaderboard/code/webdev?v=20261011",
     "checked": "2026-10-11"
    },
    {
     "id": "s4",
     "title": "LM Studio pricing",
     "url": "https://lmstudio.ai/pricing?v=20261011",
     "checked": "2026-10-11"
    },
    {
     "id": "s5",
     "title": "LM Studio app privacy policy (effective June 2026)",
     "url": "https://lmstudio.ai/app-privacy",
     "checked": "2026-10-11"
    },
    {
     "id": "s6",
     "title": "LM Studio changelog",
     "url": "https://lmstudio.ai/changelog",
     "checked": "2026-10-11"
    },
    {
     "id": "s7",
     "title": "GIGAZINE: LM Studio releases the Bionic agent app (17 Jul 2026)",
     "url": "https://gigazine.net/gsc_news/en/20260717-lm-studio-bionic",
     "checked": "2026-10-11"
    },
    {
     "id": "s8",
     "title": "arena.ai Text: Coding leaderboard",
     "url": "https://arena.ai/leaderboard/chat/text/coding?v=20261011",
     "checked": "2026-10-11"
    }
   ],
   "checked": "2026-10-11"
  },
  {
   "slug": "gemini-notebook-formerly-notebooklm",
   "url": "https://hayzlabs.com/reviews/gemini-notebook-formerly-notebooklm/",
   "name": "Gemini Notebook (formerly NotebookLM)",
   "maker": "Google",
   "kind": "app",
   "license": "proprietary",
   "runs": "cloud",
   "released": "2026-07-16",
   "status": "first-look",
   "verdict": "Worth trying",
   "verdict_line": "Free tool that answers from the files you upload, with up to 50 sources a notebook. Skip it for open web research.",
   "grades": [],
   "price": {
    "free_plan": true,
    "usd_month": 0,
    "plans": "Free with a Google account: 100 notebooks, 50 sources per notebook (s3). Google AI Plus $4.99/month raises this to 200 and 100; AI Pro $19.99/month to 500 and 300; AI Ultra from $99.99/month (s2, s3).",
    "api": null,
    "checked_on": "2026-10-11"
   },
   "how_we_rated": "First look from Google's product page, plans page, help center and launch blog post, checked 11 Oct 2026. Not rated yet. We did not run it ourselves.",
   "sources": [
    {
     "id": "s1",
     "title": "Google blog: NotebookLM is now Gemini Notebook",
     "url": "https://blog.google/innovation-and-ai/products/gemini-notebook/notebooklm-gemini-notebook/?v=20261011",
     "checked": "2026-10-11"
    },
    {
     "id": "s2",
     "title": "Google AI plans",
     "url": "https://gemini.google/subscriptions/?v=20261011",
     "checked": "2026-10-11"
    },
    {
     "id": "s3",
     "title": "Gemini Notebook Help: Upgrade Gemini Notebook",
     "url": "https://support.google.com/gemininotebook/answer/16213268?v=20261011",
     "checked": "2026-10-11"
    },
    {
     "id": "s4",
     "title": "Gemini Notebook",
     "url": "https://notebook.google/?v=20261011",
     "checked": "2026-10-11"
    }
   ],
   "checked": "2026-10-11"
  },
  {
   "slug": "gpt-live-1",
   "url": "https://hayzlabs.com/reviews/gpt-live-1/",
   "name": "GPT-Live 1",
   "maker": "OpenAI",
   "kind": "model",
   "license": "proprietary",
   "runs": "cloud",
   "released": "2026-07-08",
   "status": "graded",
   "verdict": "Worth trying",
   "verdict_line": "Top scores on hard voice agent tasks, but blind listeners prefer Gemini 3.8 Live, which also costs less.",
   "grades": [
    {
     "category": "voice",
     "score": null,
     "score_low": null,
     "score_high": null,
     "score_source": null,
     "usd_month": null,
     "checked_on": "2026-10-11"
    }
   ],
   "price": {
    "free_plan": false,
    "usd_month": null,
    "plans": "In the API: no free tier for GPT-Live sessions (s4). In ChatGPT it is the default voice on paid plans, per SiliconANGLE (s6).",
    "api": "$0.05 per minute of voice session, billed per second; backend model and tool usage billed separately (s3, s4)",
    "checked_on": "2026-10-11"
   },
   "how_we_rated": "From the Artificial Analysis Speech Agent Arena (blind votes) and Speech to Speech Index, checked 11 Oct 2026. We did not run it ourselves.",
   "sources": [
    {
     "id": "s1",
     "title": "Artificial Analysis Speech Agent Arena leaderboard",
     "url": "https://artificialanalysis.ai/speech-to-speech/arena?v=20261011",
     "checked": "2026-10-11"
    },
    {
     "id": "s2",
     "title": "Artificial Analysis speech to speech leaderboard",
     "url": "https://artificialanalysis.ai/speech-to-speech?v=20261011",
     "checked": "2026-10-11"
    },
    {
     "id": "s3",
     "title": "OpenAI API pricing",
     "url": "https://developers.openai.com/api/docs/pricing?v=20261011",
     "checked": "2026-10-11"
    },
    {
     "id": "s4",
     "title": "OpenAI API: GPT-Live 1 model page",
     "url": "https://developers.openai.com/api/docs/models/gpt-live-1?v=20261011",
     "checked": "2026-10-11"
    },
    {
     "id": "s5",
     "title": "OpenAI API: data controls",
     "url": "https://developers.openai.com/api/docs/guides/your-data?v=20261011",
     "checked": "2026-10-11"
    },
    {
     "id": "s6",
     "title": "SiliconANGLE: OpenAI launches GPT-Live voice model series",
     "url": "https://siliconangle.com/2026/07/08/openai-launches-gpt-live-voice-model-series-ahead-broad-gpt-5-6-release/",
     "checked": "2026-10-11"
    },
    {
     "id": "s7",
     "title": "The Star (Reuters): OpenAI launches GPT-Live voice models that listen and speak simultaneously",
     "url": "https://www.thestar.com.my/tech/tech-news/2026/07/09/openai-launches-gpt-live-voice-models-that-listen-and-speak-simultaneously",
     "checked": "2026-10-11"
    },
    {
     "id": "s8",
     "title": "OpenAI API changelog",
     "url": "https://developers.openai.com/api/docs/changelog?v=20261011",
     "checked": "2026-10-11"
    }
   ],
   "checked": "2026-10-11"
  },
  {
   "slug": "muse-image",
   "url": "https://hayzlabs.com/reviews/muse-image/",
   "name": "Muse Image",
   "maker": "Meta",
   "kind": "model",
   "license": "proprietary",
   "runs": "cloud",
   "released": "2026-07-07",
   "status": "graded",
   "verdict": "Use it",
   "verdict_line": "Top 10 image quality for about a cent an image, free in Meta AI. Pick Nano Banana 2.1 if you want the higher rank.",
   "grades": [
    {
     "category": "images",
     "score": 59,
     "score_low": 58,
     "score_high": 60,
     "score_source": "model",
     "score_inputs": [
      "Arena"
     ],
     "usd_month": 0,
     "checked_on": "2026-10-11"
    }
   ],
   "price": {
    "free_plan": true,
    "usd_month": 0,
    "plans": "Free in Meta AI for everyday creation; more use through Meta's subscription plans, which the launch post does not price (s5)",
    "api": "$0.01 per image on Meta's Model API page; also hosted by fal, Runway and OpenRouter (s6)",
    "checked_on": "2026-10-11"
   },
   "how_we_rated": "From the arena.ai and Artificial Analysis image leaderboards, checked 11 Oct 2026. We did not run it ourselves.",
   "sources": [
    {
     "id": "s1",
     "title": "arena.ai Text to Image leaderboard",
     "url": "https://arena.ai/leaderboard/image/text-to-image?v=20261011",
     "checked": "2026-10-11"
    },
    {
     "id": "s2",
     "title": "arena.ai Image Edit leaderboard (Single Image Edit)",
     "url": "https://arena.ai/leaderboard/image/image-edit",
     "checked": "2026-10-11"
    },
    {
     "id": "s3",
     "title": "Artificial Analysis text-to-image leaderboard",
     "url": "https://artificialanalysis.ai/image/leaderboard/text-to-image?v=20261011",
     "checked": "2026-10-11"
    },
    {
     "id": "s4",
     "title": "Artificial Analysis image editing leaderboard",
     "url": "https://artificialanalysis.ai/image/leaderboard/editing",
     "checked": "2026-10-11"
    },
    {
     "id": "s5",
     "title": "Meta newsroom: Introducing Muse Image",
     "url": "https://about.fb.com/news/2026/07/introducing-muse-image-meta-ai/",
     "checked": "2026-10-11"
    },
    {
     "id": "s6",
     "title": "Meta Model API: Muse Image",
     "url": "https://dev.meta.ai/models/muse-image",
     "checked": "2026-10-11"
    }
   ],
   "checked": "2026-10-11"
  },
  {
   "slug": "firecrawl",
   "url": "https://hayzlabs.com/reviews/firecrawl/",
   "name": "Firecrawl",
   "maker": "Firecrawl",
   "kind": "open-source",
   "license": "open-source",
   "runs": "both",
   "released": "2026-06-19",
   "status": "first-look",
   "verdict": "Worth trying",
   "verdict_line": "Popular tool that turns web pages into clean text for AI apps; free to self-host, 1,000 free API pages a month.",
   "grades": [],
   "price": {
    "free_plan": true,
    "usd_month": 0,
    "plans": "Self-host free under AGPL-3.0 (s1). API: Free 1,000 credits/month; Hobby $19/month ($16 billed yearly) for 5,000; Standard $83/month billed yearly for 100,000; Growth $333; Scale $599 (s2).",
    "api": "1 credit = 1 page on a basic scrape, crawl or map; search 2 credits per 10 results; JSON and other extraction formats +4 credits per page (s2)",
    "checked_on": "2026-10-11"
   },
   "how_we_rated": "First look from Firecrawl's GitHub page, pricing page, privacy policy and PyPI release history, checked 11 Oct 2026. Not rated yet. We did not run it ourselves.",
   "sources": [
    {
     "id": "s1",
     "title": "GitHub: firecrawl/firecrawl",
     "url": "https://github.com/firecrawl/firecrawl",
     "checked": "2026-10-11"
    },
    {
     "id": "s2",
     "title": "Firecrawl pricing",
     "url": "https://www.firecrawl.dev/pricing?v=20261011",
     "checked": "2026-10-11"
    },
    {
     "id": "s3",
     "title": "Firecrawl privacy policy",
     "url": "https://www.firecrawl.dev/privacy-policy?v=20261011",
     "checked": "2026-10-11"
    },
    {
     "id": "s4",
     "title": "PyPI: firecrawl-py",
     "url": "https://pypi.org/project/firecrawl-py/",
     "checked": "2026-10-11"
    },
    {
     "id": "s5",
     "title": "GitHub: firecrawl releases",
     "url": "https://github.com/firecrawl/firecrawl/releases",
     "checked": "2026-10-11"
    }
   ],
   "checked": "2026-10-11"
  },
  {
   "slug": "google-antigravity",
   "url": "https://hayzlabs.com/reviews/google-antigravity/",
   "name": "Google Antigravity",
   "maker": "Google",
   "kind": "dev-tool",
   "license": "proprietary",
   "runs": "cloud",
   "released": "2026-05-19",
   "status": "first-look",
   "verdict": "Worth trying",
   "verdict_line": "Free coding agent app on Gemini Flash; Claude 5.5 needs AI Pro at $19.99/month. Change the data setting first.",
   "grades": [],
   "price": {
    "free_plan": true,
    "usd_month": 0,
    "plans": "Individuals free with basic weekly limits; Google AI Pro $19.99/month and AI Ultra from $99.99/month raise limits (s1, s4); Business from $30/seat/month for eligible Gemini Enterprise customers (s1).",
    "api": null,
    "checked_on": "2026-10-11"
   },
   "how_we_rated": "First look from Google's Antigravity pages, terms and model list, the Google AI plans page and the arena.ai WebDev board, checked 11 Oct 2026. Not rated yet. We did not run it ourselves.",
   "sources": [
    {
     "id": "s1",
     "title": "Google Antigravity pricing",
     "url": "https://antigravity.google/pricing?v=20261011",
     "checked": "2026-10-11"
    },
    {
     "id": "s2",
     "title": "Google Antigravity home page",
     "url": "https://antigravity.google/?v=20261011",
     "checked": "2026-10-11"
    },
    {
     "id": "s3",
     "title": "Google Antigravity docs: Models",
     "url": "https://antigravity.google/docs/models?v=20261011",
     "checked": "2026-10-11"
    },
    {
     "id": "s4",
     "title": "Google AI plans (Gemini subscriptions)",
     "url": "https://gemini.google/subscriptions/?v=20261011",
     "checked": "2026-10-11"
    },
    {
     "id": "s5",
     "title": "Google Antigravity terms",
     "url": "https://antigravity.google/terms?v=20261011",
     "checked": "2026-10-11"
    },
    {
     "id": "s6",
     "title": "arena.ai Code Arena: WebDev leaderboard",
     "url": "https://arena.ai/leaderboard/code/webdev?v=20261011",
     "checked": "2026-10-11"
    }
   ],
   "checked": "2026-10-11"
  },
  {
   "slug": "google-antigravity-cli",
   "url": "https://hayzlabs.com/reviews/google-antigravity-cli/",
   "name": "Google Antigravity CLI",
   "maker": "Google",
   "kind": "dev-tool",
   "license": "proprietary",
   "runs": "cloud",
   "released": "2026-05-19",
   "status": "first-look",
   "verdict": "Worth trying",
   "verdict_line": "Google's free terminal coding agent, replacing Gemini CLI; turn off data use in settings if code privacy matters.",
   "grades": [],
   "price": {
    "free_plan": true,
    "usd_month": 0,
    "plans": "Individuals $0/month with basic weekly rate limits; Google AI Pro and Ultra add higher limits and an AI credit pool; business plans via Google Cloud from $30 per seat per month (s1).",
    "api": null,
    "checked_on": "2026-10-11"
   },
   "how_we_rated": "First look from Google's pricing and terms pages, the Google Developers Blog launch post and TechRadar coverage, checked 11 Oct 2026. Not rated yet. We did not run it ourselves.",
   "sources": [
    {
     "id": "s1",
     "title": "Google Antigravity pricing",
     "url": "https://antigravity.google/pricing?v=20261011",
     "checked": "2026-10-11"
    },
    {
     "id": "s2",
     "title": "Google Developers Blog: An important update, transitioning Gemini CLI to Antigravity CLI (19 May 2026)",
     "url": "https://developers.googleblog.com/an-important-update-transitioning-gemini-cli-to-antigravity-cli/",
     "checked": "2026-10-11"
    },
    {
     "id": "s3",
     "title": "Google Antigravity Additional Terms of Service",
     "url": "https://antigravity.google/terms?v=20261011",
     "checked": "2026-10-11"
    },
    {
     "id": "s4",
     "title": "TechRadar: Google is making Gemini CLI users switch to its new Antigravity 2.0 (20 May 2026)",
     "url": "https://www.techradar.com/pro/google-is-making-gemini-cli-users-switch-to-its-new-antigravity-2-0-so-what-will-it-mean-for-you",
     "checked": "2026-10-11"
    }
   ],
   "checked": "2026-10-11"
  },
  {
   "slug": "scribe-v2",
   "url": "https://hayzlabs.com/reviews/scribe-v2/",
   "name": "Scribe v2",
   "maker": "ElevenLabs",
   "kind": "model",
   "license": "proprietary",
   "runs": "cloud",
   "released": "2026-01-09",
   "status": "graded",
   "verdict": "Worth trying",
   "verdict_line": "Very accurate transcription with speaker labels and a free plan. Heavy API users pay less with MAI-Transcribe-2.",
   "grades": [
    {
     "category": "voice",
     "score": null,
     "score_low": null,
     "score_high": null,
     "score_source": null,
     "usd_month": 0,
     "checked_on": "2026-10-11"
    }
   ],
   "price": {
    "free_plan": true,
    "usd_month": 0,
    "plans": "Free plan includes speech to text, 10k credits a month; Starter $6/month (s2, s3)",
    "api": "$0.22 per hour of audio (s2)",
    "checked_on": "2026-10-11"
   },
   "how_we_rated": "From Artificial Analysis speech to text word error rates, checked 11 Oct 2026. We did not run it ourselves.",
   "sources": [
    {
     "id": "s1",
     "title": "Artificial Analysis speech to text leaderboard",
     "url": "https://artificialanalysis.ai/speech-to-text/non-streaming",
     "checked": "2026-10-11"
    },
    {
     "id": "s2",
     "title": "ElevenLabs API pricing",
     "url": "https://elevenlabs.io/pricing/api",
     "checked": "2026-10-11"
    },
    {
     "id": "s3",
     "title": "ElevenLabs pricing",
     "url": "https://elevenlabs.io/pricing",
     "checked": "2026-10-11"
    },
    {
     "id": "s4",
     "title": "ElevenLabs privacy policy",
     "url": "https://elevenlabs.io/privacy-policy",
     "checked": "2026-10-11"
    },
    {
     "id": "s5",
     "title": "ElevenLabs blog: Introducing Scribe v2",
     "url": "https://elevenlabs.io/blog/introducing-scribe-v2",
     "checked": "2026-10-11"
    },
    {
     "id": "s6",
     "title": "ElevenLabs docs: models",
     "url": "https://elevenlabs.io/docs/overview/models",
     "checked": "2026-10-11"
    }
   ],
   "checked": "2026-10-11"
  },
  {
   "slug": "claude-code",
   "url": "https://hayzlabs.com/reviews/claude-code/",
   "name": "Claude Code",
   "maker": "Anthropic",
   "kind": "dev-tool",
   "license": "proprietary",
   "runs": "both",
   "released": "2025-05-22",
   "status": "graded",
   "verdict": "Worth trying",
   "verdict_line": "Highest published score on a coding-agent test, from $20/month; OpenAI's Codex scores close for far less per task.",
   "grades": [
    {
     "category": "coding",
     "score": 70,
     "score_low": 70,
     "score_high": 70,
     "score_source": "checklist",
     "usd_month": 20,
     "checked_on": "2026-11-02"
    }
   ],
   "price": {
    "free_plan": false,
    "usd_month": 20,
    "plans": "Pro $20/month ($17/month billed yearly); Max $100/month; Ultra $200/month; Team Standard $25/seat/month; Free does not include Claude Code (s5).",
    "api": "Or pay per token with an API key (s9).",
    "checked_on": "2026-11-02"
   },
   "how_we_rated": "From the SPEC checklist, using Artificial Analysis's Coding Agent Index v1.5 results and methodology, the arena.ai WebDev board and Anthropic's pricing and data pages, checked 11 Oct 2026. We did not run it ourselves.",
   "sources": [
    {
     "id": "s1",
     "title": "Artificial Analysis: Cursor Composer 2.5 on the Coding Agent Index (20 May 2026)",
     "url": "https://artificialanalysis.ai/articles/cursor-composer-2-5-coding-agent-index",
     "checked": "2026-10-11"
    },
    {
     "id": "s2",
     "title": "Artificial Analysis: Coding agents leaderboard",
     "url": "https://artificialanalysis.ai/agents/coding-agents?v=20261011",
     "checked": "2026-10-11"
    },
    {
     "id": "s3",
     "title": "Claude Code docs: Model configuration",
     "url": "https://code.claude.com/docs/en/model-config?v=20261011",
     "checked": "2026-10-11"
    },
    {
     "id": "s4",
     "title": "arena.ai Code Arena: WebDev leaderboard",
     "url": "https://arena.ai/leaderboard/code/webdev?v=20261011",
     "checked": "2026-10-11"
    },
    {
     "id": "s5",
     "title": "Claude pricing",
     "url": "https://claude.com/pricing?v=20261102",
     "checked": "2026-11-02"
    },
    {
     "id": "s6",
     "title": "GitHub Copilot plans",
     "url": "https://github.com/features/copilot/plans?v=20261011",
     "checked": "2026-10-11"
    },
    {
     "id": "s7",
     "title": "Claude Code docs: Data usage",
     "url": "https://code.claude.com/docs/en/data-usage?v=20261011",
     "checked": "2026-10-11"
    },
    {
     "id": "s8",
     "title": "arena.ai Agent leaderboard",
     "url": "https://arena.ai/leaderboard/agent?v=20261011",
     "checked": "2026-10-11"
    },
    {
     "id": "s9",
     "title": "Claude Code docs: Overview",
     "url": "https://code.claude.com/docs/en/overview",
     "checked": "2026-10-11"
    },
    {
     "id": "s10",
     "title": "Claude Code releases on GitHub",
     "url": "https://github.com/anthropics/claude-code/releases?v=20261011",
     "checked": "2026-10-11"
    },
    {
     "id": "s11",
     "title": "Anthropic: Introducing Claude 4 (Claude Code generally available)",
     "url": "https://www.anthropic.com/news/claude-4",
     "checked": "2026-10-11"
    },
    {
     "id": "s12",
     "title": "Artificial Analysis: Claude Code vs Codex, Coding Agent Index v1.5",
     "url": "https://artificialanalysis.ai/agents/coding-agents/comparisons/claude-code-vs-codex",
     "checked": "2026-10-11"
    },
    {
     "id": "s13",
     "title": "Artificial Analysis: Coding Agent Index v1.5 methodology",
     "url": "https://artificialanalysis.ai/methodology/coding-agents-benchmarking?v=20261011",
     "checked": "2026-10-11"
    },
    {
     "id": "s14",
     "title": "Artificial Analysis: Claude Code vs OpenCode, Coding Agent Index v1.5",
     "url": "https://artificialanalysis.ai/agents/coding-agents/comparisons/claude-code-vs-opencode?v=20261011",
     "checked": "2026-10-11"
    }
   ],
   "checked": "2026-10-11"
  },
  {
   "slug": "gemini-app",
   "url": "https://hayzlabs.com/reviews/gemini-app/",
   "name": "Gemini app",
   "maker": "Google",
   "kind": "app",
   "license": "proprietary",
   "runs": "cloud",
   "released": "2024-02-08",
   "status": "graded",
   "verdict": "Depends",
   "verdict_line": "Good for writing on AI Pro with 3.8 Flash; weak at coding and hard questions, where Claude and ChatGPT score higher.",
   "grades": [
    {
     "category": "writing",
     "score": 70,
     "score_low": 70,
     "score_high": 70,
     "score_source": "checklist",
     "usd_month": 19.99,
     "checked_on": "2026-10-11"
    },
    {
     "category": "research",
     "score": 50,
     "score_low": 50,
     "score_high": 50,
     "score_source": "checklist",
     "usd_month": 19.99,
     "checked_on": "2026-10-11"
    },
    {
     "category": "coding",
     "score": 50,
     "score_low": 50,
     "score_high": 50,
     "score_source": "checklist",
     "usd_month": 19.99,
     "checked_on": "2026-10-11"
    },
    {
     "category": "agents",
     "score": 50,
     "score_low": 50,
     "score_high": 50,
     "score_source": "checklist",
     "usd_month": 19.99,
     "checked_on": "2026-10-11"
    }
   ],
   "price": {
    "free_plan": true,
    "usd_month": 19.99,
    "plans": "Free; Google AI Plus $4.99/month (3.6 Flash and 3.1 Pro); Google AI Pro $19.99/month (3.8 Flash and 3.1 Pro); Google AI Ultra $99.99 or $199.99/month. Paid plans include cloud storage (s5).",
    "api": null,
    "checked_on": "2026-10-11"
   },
   "how_we_rated": "From the SPEC checklist, scored per plan with one rule for every app: each plan is scored on the best model it lets you choose, as documented on the maker's own plan and help pages, and we name it; a default the maker names does not lower the score. Does the job is 2 when that model sits in the top band of Arena's ranks for the job, else 1; underlying quality is 2 for the top band, 1 for the next band, 0 below. The score comes from the higher-scoring plan; a tie goes to the cheaper one. Ranks are from arena.ai, and plans, limits and privacy from Google's own pages, checked 11 Oct 2026. We did not run it ourselves. Where we point to Claude, we also checked our reviews for the best non-Anthropic option and name it.",
   "sources": [
    {
     "id": "s1",
     "title": "arena.ai Text: Creative Writing leaderboard",
     "url": "https://arena.ai/leaderboard/chat/text/creative-writing?v=20261011",
     "checked": "2026-10-11"
    },
    {
     "id": "s2",
     "title": "arena.ai Text: Expert leaderboard",
     "url": "https://arena.ai/leaderboard/chat/text/expert?v=20261011",
     "checked": "2026-10-11"
    },
    {
     "id": "s3",
     "title": "arena.ai Code Arena: WebDev leaderboard",
     "url": "https://arena.ai/leaderboard/code/webdev?v=20261011",
     "checked": "2026-10-11"
    },
    {
     "id": "s4",
     "title": "arena.ai Agent leaderboard",
     "url": "https://arena.ai/leaderboard/agent?v=20261011",
     "checked": "2026-10-11"
    },
    {
     "id": "s5",
     "title": "Google AI plans (Gemini subscriptions)",
     "url": "https://gemini.google/subscriptions/?v=20261011",
     "checked": "2026-10-11"
    },
    {
     "id": "s7",
     "title": "Gemini Apps Privacy Hub",
     "url": "https://support.google.com/gemini/answer/13594961?hl=en",
     "checked": "2026-10-11"
    },
    {
     "id": "s8",
     "title": "Gemini Apps Help: usage limits",
     "url": "https://support.google.com/gemini/answer/16275805?hl=en",
     "checked": "2026-10-11"
    },
    {
     "id": "s9",
     "title": "Engadget: A native Gemini app is finally available for Windows PCs (10 Sep 2026)",
     "url": "https://engadget.com/2255197/a-native-gemini-app-is-finally-available-for-windows-pcs",
     "checked": "2026-10-11"
    },
    {
     "id": "s10",
     "title": "Android Authority: Gemini's new Windows app (10 Sep 2026)",
     "url": "https://androidauthority.com/gemini-windows-app-3710027",
     "checked": "2026-10-11"
    }
   ],
   "checked": "2026-10-11"
  },
  {
   "slug": "perplexity",
   "url": "https://hayzlabs.com/reviews/perplexity/",
   "name": "Perplexity",
   "maker": "Perplexity",
   "kind": "app",
   "license": "proprietary",
   "runs": "cloud",
   "released": "2023-03-28",
   "status": "graded",
   "verdict": "Worth trying",
   "verdict_line": "Good for cited web answers on Pro with GPT-6.1 Sol; skip it if you want free help, where Claude scores higher.",
   "grades": [
    {
     "category": "research",
     "score": 70,
     "score_low": 70,
     "score_high": 70,
     "score_source": "checklist",
     "usd_month": 17,
     "checked_on": "2026-10-18"
    }
   ],
   "price": {
    "free_plan": true,
    "usd_month": 17,
    "plans": "Free; Pro $17/month and Max $167/month, both when billed yearly, with no monthly-billing price shown; Enterprise Pro and Enterprise Max also offered (s2, s3)",
    "api": null,
    "checked_on": "2026-10-18"
   },
   "how_we_rated": "From the SPEC checklist, scored per plan with one rule for every app: each plan is scored on the best model it lets you choose, as documented on the maker's own plan and help pages, and we name it; a default the maker names does not lower the score. Does the job is 2 when that model sits in the top band of Arena's ranks for the job, else 1; underlying quality is 2 for the top band, 1 for the next band, 0 below. The score comes from the higher-scoring plan; a tie goes to the cheaper one. For research we use Arena's Search board, the better match for a search app. Ranks are from Arena's open leaderboard dataset (CC BY 4.0), and plans, models and privacy from Perplexity's own pages, checked 18 Oct 2026. We did not run it ourselves. Where we point to an Anthropic product, we also checked our reviews for the best non-Anthropic option and name it.",
   "sources": [
    {
     "id": "s1",
     "title": "Arena open leaderboard dataset: search_style_control, latest split (published 2026-10-13), CC BY 4.0",
     "url": "https://huggingface.co/datasets/lmarena-ai/leaderboard-dataset",
     "checked": "2026-10-18"
    },
    {
     "id": "s2",
     "title": "Perplexity pricing",
     "url": "https://www.perplexity.ai/hub/pricing?v=1018d",
     "checked": "2026-10-18"
    },
    {
     "id": "s3",
     "title": "Perplexity Help Center: What advanced AI models are included in my subscription?",
     "url": "https://www.perplexity.ai/help-center/en/articles/10354919-what-advanced-ai-models-are-included-in-my-subscription?v=1018d",
     "checked": "2026-10-18"
    },
    {
     "id": "s5",
     "title": "Perplexity Help Center: Data collection at Perplexity",
     "url": "https://intercom.help/perplexity-ai/en/articles/11564572-data-collection-at-perplexity",
     "checked": "2026-10-18"
    },
    {
     "id": "s6",
     "title": "Perplexity blog: Series A funding round and iPhone app launch (28 Mar 2023)",
     "url": "https://www.perplexity.ai/hub/blog/announcing-our-series-a-funding-round-and-mobile-app-launch",
     "checked": "2026-10-18"
    }
   ],
   "checked": "2026-10-18"
  },
  {
   "slug": "github-copilot",
   "url": "https://hayzlabs.com/reviews/github-copilot/",
   "name": "GitHub Copilot (including Copilot CLI and cloud agent)",
   "maker": "GitHub",
   "kind": "dev-tool",
   "license": "proprietary",
   "runs": "both",
   "released": "2022-06-21",
   "status": "first-look",
   "verdict": "Worth trying",
   "verdict_line": "Cheapest paid coding agent at $10/month, CLI on the free plan; no independent test, and personal plans train by default.",
   "grades": [],
   "price": {
    "free_plan": true,
    "usd_month": 10,
    "plans": "Free (limited to 2,000 completions and 50 chat requests; CLI and agent mode included); Pro $10/month adds the cloud agent and PR code review; Pro+ $39 adds Claude and Codex agents; Max $100 (s1).",
    "api": null,
    "checked_on": "2026-10-11"
   },
   "how_we_rated": "First look from GitHub's plans, model and data-use pages and The New Stack's report on code review benchmarks, checked 11 Oct 2026. Not rated yet. We did not run it ourselves. This page covers Copilot in the editor, Copilot CLI and the Copilot cloud agent, which are modes of one subscription.",
   "sources": [
    {
     "id": "s1",
     "title": "GitHub Copilot plans",
     "url": "https://github.com/features/copilot/plans?v=20261011",
     "checked": "2026-10-11"
    },
    {
     "id": "s2",
     "title": "GitHub Blog: Updates to GitHub Copilot interaction data usage policy (25 Mar 2026)",
     "url": "https://github.blog/news-insights/company-news/updates-to-github-copilot-interaction-data-usage-policy/?v=20261011",
     "checked": "2026-10-11"
    },
    {
     "id": "s3",
     "title": "Artificial Analysis: Cursor Composer 2.5 on the Coding Agent Index (20 May 2026)",
     "url": "https://artificialanalysis.ai/articles/cursor-composer-2-5-coding-agent-index",
     "checked": "2026-10-11"
    },
    {
     "id": "s4",
     "title": "GitHub Docs: Supported AI models in Copilot",
     "url": "https://docs.github.com/en/copilot/reference/ai-models/supported-models?v=20261011",
     "checked": "2026-10-11"
    },
    {
     "id": "s5",
     "title": "GitHub Docs: About Copilot cloud agent",
     "url": "https://docs.github.com/en/copilot/concepts/agents/cloud-agent/about-cloud-agent?v=20261011",
     "checked": "2026-10-11"
    },
    {
     "id": "s6",
     "title": "GitHub Blog: GitHub Copilot is generally available to all developers (21 Jun 2022)",
     "url": "https://github.blog/news-insights/product-news/github-copilot-is-generally-available-to-all-developers/",
     "checked": "2026-10-11"
    },
    {
     "id": "s7",
     "title": "The New Stack: Copilot tops GitHub's own AI code review benchmark. An independent one tells a different story (6 Oct 2026)",
     "url": "https://thenewstack.io/github-reviewbench-code-review/",
     "checked": "2026-10-11"
    },
    {
     "id": "s8",
     "title": "Artificial Analysis: Claude Code vs Codex, Coding Agent Index v1.5",
     "url": "https://artificialanalysis.ai/agents/coding-agents/comparisons/claude-code-vs-codex?v=20261011b",
     "checked": "2026-10-11"
    }
   ],
   "checked": "2026-10-11"
  }
 ]
}