Hayz Labs · https://hayzlabs.com/reviews/makers/openai/
makers · OpenAI
OpenAI
Everything from OpenAI we track, and our call on each. Its strongest is GPT-6 Astra for agents.
- Scored
- 6
- First looks
- 1
- On the list
- 5
- Passed on
- 0
Reviewed
Best score first; each shows its best job. First looks are new and not scored yet.
| Name | Score | Cheapest plan | Our verdict | Why |
|---|---|---|---|---|
GPT-6 AstraAI model · best at agents | #2Agents | $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. |
GPT-6.1 SolAI model · best at agents | #4Agents | $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. |
GPT Image 2.5 SunburstAI model · best at images | 100Images | Pay as you go | Worth trying | Top image model for quality and precise edits, about 21 cents a max-quality image. On a budget, use Nano Banana 2.1. |
GPT Image 2.5 FlareAI model · best at images | 92Images | Pay as you go | Depends | Second on every image board, at Sunburst's price; OpenAI says it is faster. If waits are fine, Sunburst scores higher. |
OpenAI Codex CLIOpen source · best at coding | 70Coding | $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. |
ChatGPT (GPT-6)App · best at agents | 70Agents | $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. |
OpenAI Codex CloudDeveloper tool | Not scored yet | $20/mo | First look: Worth trying | Codex tasks that keep running in OpenAI's cloud, on Plus and up; no independent test of the cloud side yet. |
GPT-Live 1AI model | No score yet | Pay as you go | Worth trying | Top scores on hard voice agent tasks, but blind listeners prefer Gemini 3.8 Live, which also costs less. |
On the list, not reviewed
Big news or good value, on our Top 20 lists.
- gpt-realtime-2.1Speech to speech model that can reason and call tools, for voice agents on OpenAI.
- gpt-realtime-2.1-miniCheaper, faster realtime voice model, about a third of the full model's audio cost.
- text-embedding-3-largeSolid default embedding model, now behind newer open and paid models on benchmarks.
- OpenAI Agents SDKOpenAI's lightweight framework for multi-agent workflows with handoffs, guardrails and tracing.
- GPT-5.6 TerraOpenAI mid-tier model, still offered as a replacement for retired chat models