Hayz Labs · https://hayzlabs.com/reviews/rankings/coding/
reviews · best for a job · 21 scored
Best for coding
Writing and fixing code, and building small apps. Ranked by score; on a tie, the cheaper one goes first. 100 is the best available today in this job.
Picked one? Coding agents for non-coders shows how to get good results from it.
New to this? What the scores and labels mean
- 0 to 100
- Our score for one job, like writing or coding, compared with everything else out today. 100 is the best available today. For AI models we build it from public data, such as Arena's blind votes and Epoch AI's independent tests; apps and tools are scored on our own checklist. The same tool can score differently for different jobs, and where two likely ranges overlap, we call it a tie.
- Our verdict
- Use it: a top choice for most people. Worth trying: good, and worth a go. Depends: good for some people or one kind of job. Wait: too early to tell, or not out for everyone yet. Skip: something else does the job better.
- First look
- Just released. We say what it is and whether to bother, and give it a score once public results exist.
- No score yet
- None of the public sources we use covers it yet, or the only evidence is from a source whose numbers we can't use, so we describe it in words instead.
- Best buy
- Nothing cheaper scores as well for that job.
- Best free
- The best score for that job that costs nothing, through a free plan or by running on your own computer.
- Pay as you go
- No monthly plan: you pay for each use, usually through a developer account.
- Checked
- The day we last confirmed the score and the price against the sources. Prices change often, so check the date.
- AI model
- The engine underneath, like GPT or Claude. You use it through an app or, if you code, directly.
- Open source
- Free to download and run on your own computer, so nothing leaves your machine.
The short answer
For coding, Claude Opus 5.5 leads Arena's WebDev board and scores 100, ahead of GPT-6 Astra at 90, and Opus 5.5 costs $4 / $20 per 1M input / output tokens against $10 / $50 for Astra, so we would not pay Astra's price.
- Best
- Claude Opus 5.5100$20/motied with 3 others
- Best free
- Claude Sonnet 5.589Free
Read our full take
Better value sits one step down: Claude Sonnet 5.5 scores 89, a range that overlaps Astra's, and GPT-6.1 Sol scores 85, both at $2 / $10. GPT-6.1 Sol costs less per Artificial Analysis Intelligence Index task: $0.72 at max effort, against $2.01 for Sonnet 5.5 at xhigh. For free use, the Claude app's free plan scores 80 on our checklist. Claude Fable 5.1 (82, $10 / $50) is not worth paying for here, since cheaper models score as high or higher. For the coding tools that run these models, Artificial Analysis's Coding Agent Index now gives a dated, independent test. Claude Code has the highest published score there but costs about $13 per task with Opus 5.5; OpenAI's Codex CLI scores close behind for about $1 per task, so it is where we would start. Both score 70 on our checklist. OpenCode, free and open source, scores 80 because it costs nothing and stores no code, though it scored lower with the model tested. GitHub Copilot ($10/month) and Cursor stay First looks until a fresh independent test covers them.
Our opinion, checked 11 Oct. The picks come straight from the scores below. How we score
This take recommends a product made by Anthropic, whose Claude models help draft our reviews. It also names the best pick from another maker.
Everything we cover for this job, best score first. Scores whose ranges overlap are tied, so choose on price and fit.
- 1Claude Opus 5.5Anthropic
Top public pick for coding, agents and hard questions at a mid price; skip it for bulk work where Sonnet 5.5 is enough.
Best for Coding and agent jobs; ranks first on the web app and agent boardsYou give up Twice Sonnet 5.5's token price; thinking can't be turned off
10089–100$20/mo - 2Claude Sonnet 5.5Anthropic
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.
Best for Everyday coding and agents with near-top results on a budgetYou give up Behind the leaders on creative writing; breaking API changes from Sonnet 5
8985–92Free - 3Claude Fable 5.1Anthropic
Very strong, but Opus 5.5 ranks higher on most boards for 40% of the price; keep Fable for cases Opus misses.
Best for Teams whose hardest long tasks still fail on Opus 5.5You give up Slower than Opus 5.5, 2.5 times its token price, and needs 30-day data retention
8776–90$100/mo - 4GPT-6.1 SolOpenAI
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.
Best for Coding and agent work near the top without top-model pricesYou give up Weak at creative writing (rank 26); not on ChatGPT Free or Go
8581–88$20/mo - 5ClaudeAnthropic
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.
Best for Free help with coding and hard questions; Pro for agent work and writingYou give up Free plan's model is weaker at creative writing; usage limits are not given in numbers
80ChecklistFree - 6LM Studio (with Bionic agent)LM Studio
Free desktop app for running open coding models privately, now with a local agent; skip it if you need frontier quality.
Best for People who want local, private AI in a desktop app, not the terminalYou give up App is closed source; slower than Ollama on small models in one test
80ChecklistFree - 7OllamaOllama
Free, private way to run open coding models like Qwen3.8 27B on your own computer; skip it if you need frontier quality.
Best for Developers who want a local model API, and private coding helpYou give up Slower than mlx-lm on a Mac in one test; its bigger models are cloud-only
80ChecklistFree - 8OpenCodeAnomaly
Free open-source coding agent for 75+ model providers; scored 54 with GLM-5.3. Skip it for one flat monthly plan.
Best for Developers who want a free agent and the choice of 75+ model providersYou give up Scores below Codex and Claude Code; several free Zen models may train on prompts
80ChecklistFree - 9GPT-6 AstraOpenAI
Second on WebDev and Agent boards, but GPT-6.1 Sol is close for a fifth of the price; middling at creative writing.
Best for The hardest coding and agent jobs, where a small lead is worth paying forYou give up Five times GPT-6.1 Sol's token price; middling at creative writing
7865–92$20/mo - 10ChatGPT (GPT-6)OpenAI
Good for agent work and coding on Plus with GPT-6 Astra. For writing, Gemini AI Pro and Claude Pro score higher.
Best for People who want one app for agent tasks and coding on the $20 Plus planYou give up Writing scores 50
70Checklist$20/mo - 11Claude CodeAnthropic
Highest published score on a coding-agent test, from $20/month; OpenAI's Codex scores close for far less per task.
Best for Developers who already pay for Claude and want the strongest current coding modelYou give up No free plan; costs far more per task than Codex, which scores close
70Checklist$20/mo - 12OpenAI Codex CLIOpenAI
Close to the top coding-agent score at a fraction of the cost per task; needs ChatGPT Plus or an API key.
Best for ChatGPT Plus subscribers who want a terminal coding agent at no extra costYou give up Default model unnamed; Plus limits vary widely and personal plans train by default
70Checklist$20/mo - 13Gemini 4 ArgonGoogle
Google's top model, first in Arena's text votes, now in Google AI Plus from $4.99/month; votes are still early.
Best for Everyday chat and writing at the top of Arena votes, from $4.99 a monthYou give up Votes still early; ranks 10th for web app coding, and API price doubles on 1 Dec
6461–68$4.99/mo - 14Muse Spark 1.3Meta
Solid coding and agent model at a low token price; GPT-6.1 Sol and Claude rank higher if quality matters most.
Best for Low-cost coding and agent work through Meta's APIYou give up Ranks well below GPT-6.1 Sol on agent tasks; the cheapest tier uses your data
6058–62$5/mo - 15Grok 4.7xAI
Decent agent and web-app work, weak at writing; GPT-6.1 Sol ranks higher for less per task.
5552–57$30/mo - 16MiMo-V2.6-ProXiaomi
Top open-weights model on the Artificial Analysis index, at cents per task; closed leaders still rank well above it.
Best for Cheap coding and agent work, or an MIT model to host yourselfYou give up Self-hosting needs 8 to 16 GPUs; data on Xiaomi's platform is stored in China
5250–54Free - 17Gemini appGoogle
Good for writing on AI Pro with 3.8 Flash; weak at coding and hard questions, where Claude and ChatGPT score higher.
Best for People who already pay Google for storage and want a writing helperYou give up Free plan runs a much weaker model; usage limits are not published in numbers
50Checklist$19.99/mo - 18GLM-5.3-FlashZ.ai (Zhipu)
Very cheap MIT model, 13th on arena.ai Math; for coding and agents MiMo-V2.6-Pro ranks higher for less per task.
Best for Cheap maths-heavy or bulk work through an API or your own serversYou give up Mid-table for coding and weak at creative writing; hosting needs a server
4847–49Free - 19Claude Haiku 5.5Anthropic
Very cheap small model for summaries, sorting and coding subagents; use Sonnet 5.5 for hard coding work.
4442–46Pay as you go - 20Gemini 3.8 FlashGoogle
Low-cost Google model near the top for writing and maths; rankings are preliminary and API prices double in January.
Best for Writing, maths and video or audio input at a low token priceYou give up Weak at web app coding (rank 31); API prices double on 1 Jan 2027
4342–45$19.99/mo - 21Mistral Large 4Mistral AI
Mistral's 1T preview scores no higher than 72 in any job; weights due 31 Oct. Use GLM-5.3 or Kimi K3 now.
3634–38Pay as you go
Score against cost
Up is a better score, left is cheaper. Plans are the cheapest that unlock the scored feature. 2 without a monthly price are left off; see Pay per use.
Coding: free gets you a score of 89. About $20 a month gets you 100; paying more buys no better score.
Names in one box share a score band and a price step; each shows its own score, best first.
Best value: Claude Sonnet 5.5 (89, free) · Claude Opus 5.5 (100, $20/mo)
Not on the map, as they have no monthly price: Claude Haiku 5.5 (44), Mistral Large 4 (36).
Show as a table
| Name | Score | Cheapest plan | Price step | Best value |
|---|---|---|---|---|
| Claude Opus 5.5 | 100 · likely 89–100 | $20 a month | $10 to $29 | Yes |
| Claude Sonnet 5.5 | 89 · likely 85–92 | Free | Free | Yes |
| Claude Fable 5.1 | 87 · likely 76–90 | $100 a month | $100 or more | |
| GPT-6.1 Sol | 85 · likely 81–88 | $20 a month | $10 to $29 | |
| Claude | 80 | Free | Free | |
| LM Studio (with Bionic agent) | 80 | Free | Free | |
| Ollama · open source | 80 | Free | Free | |
| OpenCode · open source | 80 | Free | Free | |
| GPT-6 Astra | 78 · likely 65–92 | $20 a month | $10 to $29 | |
| ChatGPT (GPT-6) | 70 | $20 a month | $10 to $29 | |
| Claude Code | 70 | $20 a month | $10 to $29 | |
| OpenAI Codex CLI · open source | 70 | $20 a month | $10 to $29 | |
| Gemini 4 Argon | 64 · likely 61–68 | $4.99 a month | Under $10 | |
| Muse Spark 1.3 | 60 · likely 58–62 | $5 a month | Under $10 | |
| Grok 4.7 | 55 · likely 52–57 | $30 a month | $30 to $99 | |
| MiMo-V2.6-Pro · open source | 52 · likely 50–54 | Free | Free | |
| Gemini app | 50 | $19.99 a month | $10 to $29 | |
| GLM-5.3-Flash · open source | 48 · likely 47–49 | Free | Free | |
| Claude Haiku 5.5 | 44 · likely 42–46 | No monthly price | ||
| Gemini 3.8 Flash | 43 · likely 42–45 | $19.99 a month | $10 to $29 | |
| Mistral Large 4 · open source | 36 · likely 34–38 | No monthly price |
Up is a better score, left is cheaper to run. Cost to run one typical agent task (arena.ai, middle of the range), for the 11 scored here that have one; the others are left off rather than mixed with figures from another test.
| # | Name | Score | Source | Cheapest plan | Our verdict | Why |
|---|---|---|---|---|---|---|
| 1 | Claude Opus 5.5Anthropic · AI model | 100likely 89–100Tied with #2, #3, #9 | Our score | $20/mo | Use itBest buy | Top public pick for coding, agents and hard questions at a mid price; skip it for bulk work where Sonnet 5.5 is enough. |
| 2 | Claude Sonnet 5.5Anthropic · AI model | 89likely 85–92Tied with 4 others | Our score | Free | Use itBest buyBest free | 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. |
| 3 | Claude Fable 5.1Anthropic · AI model | 87likely 76–90Tied with 4 others | Our score | $100/mo | Skip | Very strong, but Opus 5.5 ranks higher on most boards for 40% of the price; keep Fable for cases Opus misses. |
| 4 | GPT-6.1 SolOpenAI · AI model | 85likely 81–88Tied with #2, #3, #9 | Our score | $20/mo | Use it | 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. |
| 5 | ClaudeAnthropic · App | 80Tied with #6, #7, #8 | Our checklist | Free | Worth trying | 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. |
| 6 | LM Studio (with Bionic agent)LM Studio · App | 80Tied with #5, #7, #8 | Our checklist | Free | Worth tryingBest on your computer | Free desktop app for running open coding models privately, now with a local agent; skip it if you need frontier quality. |
| 7 | OllamaOllama · Open source | 80Tied with #5, #6, #8 | Our checklist | Free | Worth trying | Free, private way to run open coding models like Qwen3.8 27B on your own computer; skip it if you need frontier quality. |
| 8 | OpenCodeAnomaly · Open source | 80Tied with #5, #6, #7 | Our checklist | Free | Worth trying | Free open-source coding agent for 75+ model providers; scored 54 with GLM-5.3. Skip it for one flat monthly plan. |
| 9 | GPT-6 AstraOpenAI · AI model | 78likely 65–92Tied with 5 others | Our score | $20/mo | Depends | Second on WebDev and Agent boards, but GPT-6.1 Sol is close for a fifth of the price; middling at creative writing. |
| 10 | ChatGPT (GPT-6)OpenAI · App | 70Tied with #11, #12 | Our checklist | $20/mo | Depends | Good for agent work and coding on Plus with GPT-6 Astra. For writing, Gemini AI Pro and Claude Pro score higher. |
| 11 | Claude CodeAnthropic · Developer tool | 70Tied with #10, #12 | Our checklist | $20/mo | Worth trying | Highest published score on a coding-agent test, from $20/month; OpenAI's Codex scores close for far less per task. |
| 12 | OpenAI Codex CLIOpenAI · Open source | 70Tied with #10, #11 | Our checklist | $20/mo | Worth trying | Close to the top coding-agent score at a fraction of the cost per task; needs ChatGPT Plus or an API key. |
| 13 | Gemini 4 ArgonGoogle · AI model | 64likely 61–68Tied with #9, #14 | Our score | $4.99/mo | Use it | Google's top model, first in Arena's text votes, now in Google AI Plus from $4.99/month; votes are still early. |
| 14 | Muse Spark 1.3Meta · AI model | 60likely 58–62Tied with #13 | Our score | $5/mo | Worth trying | Solid coding and agent model at a low token price; GPT-6.1 Sol and Claude rank higher if quality matters most. |
| 15 | Grok 4.7xAI · AI model | 55likely 52–57Tied with #16 | Our score | $30/mo | Skip | Decent agent and web-app work, weak at writing; GPT-6.1 Sol ranks higher for less per task. |
| 16 | MiMo-V2.6-ProXiaomi · Open source | 52likely 50–54Tied with #15 | Our score | Free | Worth trying | Top open-weights model on the Artificial Analysis index, at cents per task; closed leaders still rank well above it. |
| 17 | Gemini appGoogle · App | 50 | Our checklist | $19.99/mo | Depends | Good for writing on AI Pro with 3.8 Flash; weak at coding and hard questions, where Claude and ChatGPT score higher. |
| 18 | GLM-5.3-FlashZ.ai (Zhipu) · Open source | 48likely 47–49 | Our score | Free | Worth trying | Very cheap MIT model, 13th on arena.ai Math; for coding and agents MiMo-V2.6-Pro ranks higher for less per task. |
| 19 | Claude Haiku 5.5Anthropic · AI model | 44likely 42–46Tied with #20 | Our score | Pay as you go | Worth trying | Very cheap small model for summaries, sorting and coding subagents; use Sonnet 5.5 for hard coding work. |
| 20 | Gemini 3.8 FlashGoogle · AI model | 43likely 42–45Tied with #19 | Our score | $19.99/mo | Worth trying | Low-cost Google model near the top for writing and maths; rankings are preliminary and API prices double in January. |
| 21 | Mistral Large 4Mistral AI · AI model | 36likely 34–38 | Our score | Pay as you go | Wait | Mistral's 1T preview scores no higher than 72 in any job; weights due 31 Oct. Use GLM-5.3 or Kimi K3 now. |
Nothing matches all of that. , or try everything.
What changed
New leaders and rating changes in coding; our working is under Details. All news
- Worse: ChatGPT (GPT-6): coding rating moved down.
Details
Audit under our rule that each plan is rated on the best model it includes: OpenAI names GPT-6 Sol for Plus chat and GPT-6 Luna for Free, so both checklist lines now use those models instead of the best model in Plus. Coding fell because GPT-6 Sol ranks lower after the Text Coding check, so Free now scores higher. Research and writing keep their ratings but now come from Free. Dropped the Meta Muse test and Instruction Following board, and stopped counting the 24 Sep Tom's Guide test, which did not name the ChatGPT-6 version.
- Better: ChatGPT (GPT-6): coding rating moved up.
Details
Rule change: each plan is scored on the best model it lets you choose. OpenAI's help page says Plus includes GPT-6 Astra in ChatGPT Work and Codex, while Plus chat offers only GPT-6 Sol. Agents and coding on Plus are now rated on Astra, so coding returns to 7/10. Research and writing stay on chat models, so they are unchanged. Arena text URLs moved to the /leaderboard/chat/text/ form.
- Worse: Claude Sonnet 5.5: coding rating moved down.
Details
Audit: re-derived the ratings with our rule of counting each model once on a leaderboard. Coding moved down a step (Text Coding is 16th of distinct models, below its WebDev rank). Writing moved up a step (Instruction Following is 10th of distinct models). best_category now agents. Agent cost relabelled as the 14-day median; PCWorld cited for the free-plan model.