GPT-6 Astra at a glance
GPT-6 Astra launched in the API with a 1M-token context, max reasoning effort and new agent controls such as async tool calling.
Current standing
As of
- Rank
- #3of 43 models
GPT-6 Astra (Max) · 56.9 on the Artificial Analysis Intelligence Index v5.0.1 (opens in a new tab)
- Recorded at publication
- 57 (v5.0.1)
as of
- Current leader
- Gemini 4 Carbon (Google) · 60
Computed from our daily snapshot; it changes as new models are measured. OpenAI releases GPT-6 Astra, its most capable model for end-to-end work
What changed with GPT-6 Astra
OpenAI released GPT-6 Astra (gpt-6-astra) on September 3, 2026, calling it its most capable model for the hardest end-to-end work: reasoning, coding, computer use, research and document creation. It has a context window of about 1.05M tokens, up to 128K output tokens and reasoning effort levels up to 'max'.
The same day OpenAI added async tool calling, mid-turn steering and mid-conversation reasoning effort changes. OpenAI also runs misalignment monitoring on supported Astra requests. Astra does not support the 'none' reasoning effort, custom temperature/top_p or logprobs, and tool calling requires the Responses API.
GPT-6 Astra benchmarks and pricing
Values are copied from the linked source on the date shown. They are not IA Compa measurements.
- Artificial Analysis Intelligence Index (max)as of · Source for Artificial Analysis Intelligence Index (max): artificialanalysis.ai (opens in a new tab)
- 57 (v5.0.1)
- Output speed (max)as of · Source for Output speed (max): artificialanalysis.ai (opens in a new tab)
- 43 tokens/s
- Input price per 1M tokensas of · Source for Input price per 1M tokens: developers.openai.com (opens in a new tab)
- $10.00
- Output price per 1M tokensas of · Source for Output price per 1M tokens: developers.openai.com (opens in a new tab)
- $50.00
- Context windowas of · Source for Context window: developers.openai.com (opens in a new tab)
- 1.05M tokens
How it compares
Other models in the same category that we have covered, using each model's latest recorded value. (Models)
GPT-6 Astra reasoning effort
Artificial Analysis tests this model at several reasoning-effort settings. Higher effort usually scores higher but costs more and answers more slowly.
GPT-6 Astra by effort level
Higher effort lets the model think longer. Here is what each setting scores, how fast it writes, and what it costs on Artificial Analysis.
| Effort | Intelligence Index | Output speed (tokens per second) | Cost to run the index | Blended price (3:1) (per 1M tokens) | Price in / out (per 1M tokens) | First token |
|---|---|---|---|---|---|---|
| Max (GPT-6 Astra (max)) · Best: highest effort with a score, used in the leaderboard | 57 | 43 | $5,661 | $20 | $10 / $50 | 238 s |
| Extra high (GPT-6 Astra (xhigh)) | 55 | 43 | $3,746 | $20 | $10 / $50 | 125 s |
| High (GPT-6 Astra (high)) | 55 | 44 | $3,204 | $20 | $10 / $50 | 57 s |
| Medium (GPT-6 Astra (medium)) | 52 | 47 | $2,638 | $20 | $10 / $50 | 7.1 s |
| Low (GPT-6 Astra (low)) | 43 | 44 | $1,651 | $20 | $10 / $50 | 4.2 s |
Benchmarked by AA: max · xhigh · high · medium · low of 6 available
Where to compare GPT-6 Astra
Independent leaderboards and official resources for comparing this model yourself. Rankings change often; each site shows its own current data.
Leaderboards
General leaderboards
Why GPT-6 Astra matters for businesses
At $10 input and $50 output per million tokens, Astra is OpenAI's premium tier; Artificial Analysis scores it 57 on its Intelligence Index v5.0.1, the highest among the OpenAI models we checked, but at a measured 43 tokens/s. For most business automations, Sol or Luna will be cheaper; Astra fits complex, high-value tasks where accuracy matters most.
Sources for this story
- OpenAI API changelog (opens in a new tab)Official
OpenAI · developers.openai.com
- GPT-6 Astra model page (opens in a new tab)Official
OpenAI · developers.openai.com
- Using the GPT-6 model family (latest model guide) (opens in a new tab)Official
OpenAI · developers.openai.com
- GPT-6 Astra model page (opens in a new tab)Benchmark
Artificial Analysis · artificialanalysis.ai
Last verified: . IA Compa summarizes public announcements for business readers. We are not affiliated with the companies mentioned. Always confirm details, pricing, and availability with the official source.
- #llm
- #openai
- #gpt-6
- #frontier-model
- #agents
