GPT-6 Astra
$35.00
estimated usage cost
1,000 requests × (1,000 uncached input × $10.00 + 500 output × $50.00) / 1,000,000 = $35.00
base rates: up to 272,000 input tokens per request, inclusive.
Text generation / OpenAI
GPT-6 Astra
GPT-6 Astra is an OpenAI reasoning model for demanding coding, research, computer use, and document creation. It is designed to work through complex tasks from an initial request to a completed result. Its published context window is 1,050,000 tokens. The advertised input rate is $10.00 per million input tokens. The advertised output rate is $50.00 per million output tokens. Additional pricing brackets apply when the input exceeds a published context threshold.
api id: openai/gpt-6-astra
how much does gpt-6 astra cost?
published base api rates · usd per 1m tokens
- input
- $10.00
- output
- $50.00
- cache read
- $1.00
- cache write · default ttl
- $12.50
| total input | input | output | cache read | cache write |
|---|---|---|---|---|
| over 272,000 tokens | $20.00 | $75.00 | $2.00 | $25.00 |
crossing a threshold applies that bracket to the whole request. cache creation can have a separate charge.
rates are published by the impossibl api. balance-purchase fees, taxes, tool charges, and workspace-specific pricing are separate.
specifications & capabilities
- creator
- OpenAI
- model type
- Text generation
- context window
- 1.05m tokens
- maximum output
- not published in the catalog
- input types
- text, image
- output types
- text
- tool calling
- not verified in the catalog
unknown means the public catalog does not establish that fact. a model's documented capabilities can differ from what a particular api supports.
estimate your usage
estimate your api cost
estimate text token usage for your workload. all amounts are in usd.
estimates use published impossibl api usage rates. configured billing adjustments, funding fees, taxes, and custom workspace rates are excluded and may change the final amount.
image and audio usage, cache creation, and additional tool charges are excluded. reasoning tokens count as billable output, even when they are not visible in the response.
compare more Text generation models
- GPT Chat Latest$5.00 input / 1m tokenscompare →
- GPT OSS 120B$0.35 input / 1m tokenscompare →
- GPT OSS 20B$0.075 input / 1m tokenscompare →
- GPT-3.5 Turbo$0.50 input / 1m tokenscompare →
independent benchmarks
Results from Artificial Analysis. Each configuration is listed separately. These evaluations describe the tested model configuration; performance and costs can differ on Impossibl. Last checked 2026-09-10.
GPT-6 Astra (max)
| Benchmark | Score |
|---|---|
| Intelligence Index | 52.81 |
| Omniscience Index | 43.40 |
| GPQA | 96.06% |
| Humanity’s Last Exam | 54.68% |
| SciCode | 56.48% |
| CritPt | 31.71% |
| Terminal-Bench v2.1 | 88.39% |
| Terminal-Bench v4.0 | 59.09% |
| AA-LCR | 80.67% |
| Multilingual LCR | 35.00% |
| GDPval Elo | 1580.20 |
| GDP.pdf All-pass | 31.00% |
| AutomationBench Partial | 68.49% |
| AnalystAgent | 51.25% |
| MMMU-Pro | 86.88% |
| Tau Banking | 41.44% |
| coding index (API) | 76.9 |
| agentic index (API) | 51.5 |
| intelligence index (API) | 52.8 |
Unmeasured scores are omitted. API values retain their source units.
view source ↗GPT-6 Astra (high)
| Benchmark | Score |
|---|---|
| Intelligence Index | 51.05 |
| Omniscience Index | 43.73 |
| GPQA | 94.95% |
| Humanity’s Last Exam | 53.06% |
| SciCode | 55.44% |
| CritPt | 28.86% |
| Terminal-Bench v2.1 | 89.89% |
| Terminal-Bench v4.0 | 54.04% |
| AA-LCR | 80.00% |
| GDPval Elo | 1529.70 |
| GDP.pdf All-pass | 31.00% |
| AutomationBench Partial | 66.61% |
| MMMU-Pro | 86.42% |
| Tau Banking | 40.00% |
| coding index (API) | 77.1 |
| agentic index (API) | 48.7 |
| intelligence index (API) | 51 |
Unmeasured scores are omitted. API values retain their source units.
view source ↗GPT-6 Astra (low)
| Benchmark | Score |
|---|---|
| Intelligence Index | 45.99 |
| Omniscience Index | 40.55 |
| GPQA | 93.13% |
| Humanity’s Last Exam | 49.21% |
| SciCode | 54.05% |
| CritPt | 26.29% |
| Terminal-Bench v2.1 | 88.01% |
| Terminal-Bench v4.0 | 41.92% |
| AA-LCR | 80.00% |
| GDPval Elo | 1419.29 |
| GDP.pdf All-pass | 30.40% |
| AutomationBench Partial | 59.10% |
| MMMU-Pro | 84.62% |
| Tau Banking | 31.96% |
| coding index (API) | 75.7 |
| agentic index (API) | 39.5 |
| intelligence index (API) | 46 |
Unmeasured scores are omitted. API values retain their source units.
view source ↗GPT-6 Astra (medium)
| Benchmark | Score |
|---|---|
| Intelligence Index | 49.67 |
| Omniscience Index | 42.22 |
| GPQA | 93.94% |
| Humanity’s Last Exam | 52.73% |
| SciCode | 54.17% |
| CritPt | 29.14% |
| Terminal-Bench v2.1 | 89.51% |
| Terminal-Bench v4.0 | 49.49% |
| AA-LCR | 79.67% |
| GDPval Elo | 1500.67 |
| GDP.pdf All-pass | 30.40% |
| AutomationBench Partial | 64.59% |
| MMMU-Pro | 85.09% |
| Tau Banking | 35.46% |
| coding index (API) | 76.7 |
| agentic index (API) | 46.3 |
| intelligence index (API) | 49.7 |
Unmeasured scores are omitted. API values retain their source units.
view source ↗GPT-6 Astra (Non-reasoning)
| Benchmark | Score |
|---|---|
| Intelligence Index | 45.17 |
| Omniscience Index | 26.55 |
| GPQA | 89.49% |
| Humanity’s Last Exam | 37.12% |
| SciCode | 53.47% |
| CritPt | 19.71% |
| Terminal-Bench v2.1 | 89.14% |
| Terminal-Bench v4.0 | 50.51% |
| AA-LCR | 70.67% |
| GDPval Elo | 1542.59 |
| GDP.pdf All-pass | 26.80% |
| AutomationBench Partial | 61.77% |
| MMMU-Pro | 82.02% |
| Tau Banking | 37.11% |
| coding index (API) | 76.2 |
| agentic index (API) | 47.8 |
| intelligence index (API) | 45.2 |
Unmeasured scores are omitted. API values retain their source units.
view source ↗GPT-6 Astra (xhigh)
| Benchmark | Score |
|---|---|
| Intelligence Index | 52.51 |
| Omniscience Index | 43.42 |
| GPQA | 96.26% |
| Humanity’s Last Exam | 54.59% |
| SciCode | 55.67% |
| CritPt | 31.43% |
| Terminal-Bench v2.1 | 89.14% |
| Terminal-Bench v4.0 | 59.60% |
| AA-LCR | 80.00% |
| GDPval Elo | 1555.23 |
| GDP.pdf All-pass | 32.20% |
| AutomationBench Partial | 67.18% |
| MMMU-Pro | 86.24% |
| Tau Banking | 43.09% |
| coding index (API) | 75.9 |
| agentic index (API) | 50.6 |
| intelligence index (API) | 52.5 |
Unmeasured scores are omitted. API values retain their source units.
view source ↗frequently asked questions
What is GPT-6 Astra?
GPT-6 Astra is an OpenAI reasoning model for demanding coding, research, computer use, and document creation. It is designed to work through complex tasks from an initial request to a completed result.
How much does GPT-6 Astra cost?
GPT-6 Astra costs $10.00 per million input tokens and $50.00 per million output tokens through the Impossibl API. Cache reads cost $1.00 per million tokens; cache writes cost $12.50 per million tokens. These base rates apply through 272,000 total input tokens per request. Above 272,000 input tokens, the long-context rates in the pricing table apply to the whole request. All rates are in USD.
What is the context length of GPT-6 Astra?
GPT-6 Astra has a 1,050,000-token context window through the Impossibl API.
What inputs and outputs does GPT-6 Astra support?
GPT-6 Astra accepts text and image as input. GPT-6 Astra returns text.
How do I use GPT-6 Astra through an API?
GPT-6 Astra is available through the Impossibl API. Use openai/gpt-6-astra as the model identifier with /v1/chat/completions. The quickstart explains how to create an API key and send your first request.
When was GPT-6 Astra released?
OpenAI released GPT-6 Astra on September 3, 2026.
sources & coverage
this page describes GPT-6 Astra as listed by the impossibl public api. prices and availability come from that catalog, refreshed here every five minutes. this is an impossibl offering, not a comparison of independent hosting-provider quotes.
model descriptions and release dates, where available, use the creator sources linked in the faq. api prices and limits use the impossibl catalog. benchmark results, where available, are credited to Artificial Analysis in the benchmarks section. report a correction →