GPT-6 Astra
OpenAIReleased Sep 3, 2026Closed weightsgpt-6-astra
- Intelligence (Epoch)
- 169 · 2nd of 111
- Knowledge cutoff
- April 2026
- Context
- 1M tokens
- Max output
- 128K tokens
- Input
- text, image
- OpenAI's API
- yes, as gpt-6-astra
Read Oct 17, 00:00 UTC from the hosts' published offers and OpenRouter, spec sheet from OpenAI · cutoff: platform.openai.com
Intelligence for the price
On the best-deal line: no model scores higher for less than $60.00 per M.
Epoch Capabilities Index against the lowest input + output price per M, log scale · prices Oct 17. Score: Epoch AI, CC BY 4.0, 169 with an interval of 165 to 175.
Against its rivals
The best of its price range and the top score
- GPT-6 Astrathis model$60.00169
- Claude Opus 5.5top score$24.00169
- Claude Fable 5.1same price range$60.00166
- Claude Opus 5same price range$30.00164
Same level for 5× less: GPT-6 Sol at OpenAI, $12.00 per M input + output · their Epoch intervals overlap (164 against 169).
Bars: Epoch Capabilities Index (Epoch AI, CC BY 4.0), drawn from 100, below the lowest score in the catalogue · price: lowest input + output per M.
| Provider | Input | Output | Cache ↑ | Checked |
|---|---|---|---|---|
| OpenAI | $10.00 | $50.00 | $1.00 | Oct 17 |
1 provider · prices per million tokens, read Oct 17, 00:00 UTC
| Provider | Weights | Context | Max output | Tools | Structured outputs | Reasoning |
|---|---|---|---|---|---|---|
| OpenAI | — | 1M | 125K | — | — | — |
As served by each host and declared to OpenRouter; not verified by us · weights: the precision the host declares
GPT-6 Astra spec sheet, field by field, with sources
GPT-6 Astra spec sheetmaker-declared
What OpenAI declares about the model itself — read from its own weights repository and documentation, not from any host. What each endpoint actually serves is in the pricing above: the two can differ, and both are true.
- Context window
- 1,050,000 tokens
- Parameters
- not published by the maker
- Active experts
- not published by the maker
- Weight precision
- not applicable — no weights publishedcurated
- License
- Proprietary — API onlycuratedScope: the licence on the published weights, and nothing else. It is not the maker's acceptable-use policy, which is a separate document, and it is not the contract you sign with whichever host you call — hosts set their own terms, which this index does not read.
- Knowledge cutoff
- April 30, 2026
- Modalities
- text · image → text
- Max output (official)
- 128,000 tokens
- Release date
- not published by the maker
- Maker lifecycle status
- to verify
developers.openai.com · maker's documentation · read 2026-09-19 · 5 of 10 fields published by the maker
| Provider | Chatbot 1,000 in / 500 out | RAG / long context 8,000 in / 500 out | Batch summarize 4,000 in / 1,000 out |
|---|---|---|---|
| OpenAI | $35.00 | $105.00 | $90.00 |
What each host bills for 1,000 requests of a representative job
Your own workload
Other OpenAI models · all its models, hosts and history ›
- gpt-oss-120b19 providers · from $0.18
- gpt-oss-20b13 providers · from $0.12
- GPT-5.6 Luna3 providers · from $1.40
- GPT-5.6 Sol3 providers · from $24.00
- GPT-5.6 Terra3 providers · from $14.00
- GPT-4.11 provider · from $10.00
Head-to-head: GPT-6 Astra (OpenAI) vs Claude Fable 5 (Anthropic) · Grok 4.6 (xAI) vs GPT-6 Astra (OpenAI) · DeepSeek V4 Pro 0813 (DeepSeek) vs GPT-6 Astra (OpenAI)