GPT 6 Astra
GPT-6 Astra is OpenAI's most capable model for complex reasoning, coding, computer use, research, and document creation.
text + image + file → text · made by OpenAI
$10→$50per Mtok in / out
Databricks · 14 sellers
Sold by 33 ways
| Seller | Lane | Rate |
|---|---|---|
| Databricks cloud | standard | $10per Mtok in$50per Mtok out$1per Mtok cached$12.5per Mtok cache writeraw.githubusercontent.com · read 2026-11-01 |
| DigitalOcean cloud | standard | $10per Mtok in$50per Mtok out$1per Mtok cached$12.5per Mtok cache writemodels.dev · read 2026-11-01 |
| Microsoft Azure AI cloud | standard | $10per Mtok in$50per Mtok out$1per Mtok cached$12.5per Mtok cache writemodels.dev · read 2026-11-01 |
| AWS Bedrock cloud | standard | $11per Mtok in$55per Mtok out$1.1per Mtok cached$13.75per Mtok cache writemodels.dev · read 2026-11-01 |
| Databricks cloud | priority | $20per Mtok in$100per Mtok out$2per Mtok cachedraw.githubusercontent.com · read 2026-11-01 |
| Nous Research 3 ways | $10per Mtok in$50per Mtok out$1per Mtok cached$12.5per Mtok cache write | |
| Nous Research aggregator | batch | $5per Mtok in$25per Mtok out$0.5per Mtok cached$6.25per Mtok cache writeinference-api.nousresearch.com · read 2026-09-20 |
| Nous Research aggregator | standard | $10per Mtok in$50per Mtok out$1per Mtok cached$12.5per Mtok cache writeinference-api.nousresearch.com · read 2026-09-20 |
| Nous Research aggregator | fast | $20per Mtok in$100per Mtok out$2per Mtok cached$25per Mtok cache writeinference-api.nousresearch.com · read 2026-09-20 |
| OpenRouter 10 ways | $10per Mtok in$50per Mtok out$1per Mtok cached$12.5per Mtok cache write | |
| OpenRouter aggregator | batch | $5per Mtok in$25per Mtok out$0.5per Mtok cached$6.25per Mtok cache writeopenrouter.ai · read 2026-09-20 |
| OpenRouter aggregator | openai/flex | $5per Mtok in$25per Mtok out$0.5per Mtok cached$6.25per Mtok cache writeopenrouter.ai · read 2026-09-20 |
| OpenRouter aggregator | standard | $10per Mtok in$50per Mtok out$1per Mtok cached$12.5per Mtok cache writeopenrouter.ai · read 2026-09-20 |
| OpenRouter aggregator | amazon-bedrock/global | $10per Mtok in$50per Mtok out$1per Mtok cached$12.5per Mtok cache writeopenrouter.ai · read 2026-10-30 |
| OpenRouter aggregator | azure | $10per Mtok in$50per Mtok out$1per Mtok cached$12.5per Mtok cache writeopenrouter.ai · read 2026-09-20 |
| OpenRouter aggregator | openai | $10per Mtok in$50per Mtok out$1per Mtok cached$12.5per Mtok cache writeopenrouter.ai · read 2026-09-20 |
| OpenRouter aggregator | amazon-bedrock/us-west-2 | $11per Mtok in$55per Mtok out$1.1per Mtok cached$13.75per Mtok cache writeopenrouter.ai · read 2026-09-23 |
| OpenRouter aggregator | azure/us | $11per Mtok in$55per Mtok out$1.1per Mtok cached$13.75per Mtok cache writeopenrouter.ai · read 2026-09-20 |
| OpenRouter aggregator | openai/fast | $20per Mtok in$100per Mtok out$2per Mtok cached$25per Mtok cache writeopenrouter.ai · read 2026-09-20 |
| OpenRouter aggregator | openai/ultrafast | $60per Mtok in$300per Mtok out$6per Mtok cached$75per Mtok cache writeopenrouter.ai · read 2026-09-30 |
| AtlasCloud aggregator | standard | $10per Mtok in$50per Mtok out$1per Mtok cachedapi.atlascloud.ai · read 2026-09-20 |
| ElectronHub aggregator | standard | $10per Mtok in$50per Mtok outapi.electronhub.ai · read 2026-11-01 |
| GitHub Copilot aggregator | standard | $10per Mtok in$50per Mtok out$1per Mtok cached$12.5per Mtok cache writemodels.dev · read 2026-11-01 |
| Inference.net aggregator | standard | $10per Mtok in$50per Mtok out$1per Mtok cachedapi.inference.net · read 2026-11-01 |
| OpenAI 0 ways | $10per Mtok in$50per Mtok out$1per Mtok cached | |
| OpenAI aggregatornot on the seller's list today | standard | $10per Mtok in$50per Mtok out$1per Mtok cachedmodels.dev · read 2026-09-05 |
| RedPill aggregator | standard | $10per Mtok in$50per Mtok out$1per Mtok cached$12.5per Mtok cache writeapi.redpill.ai · read 2026-10-28 |
| Requesty aggregator | openai | $10per Mtok in$50per Mtok out$1per Mtok cachedrouter.requesty.ai · read 2026-09-20 |
| Venice AI aggregator | standard | $10per Mtok in$50per Mtok out$1per Mtok cached$12.5per Mtok cache writemodels.dev · read 2026-11-01 |
| Vercel AI Gateway 2 ways | $10per Mtok in$50per Mtok out$1per Mtok cached$12.5per Mtok cache write | |
| Vercel AI Gateway aggregator | standard | $10per Mtok in$50per Mtok out$1per Mtok cached$12.5per Mtok cache writeai-gateway.vercel.sh · read 2026-09-20 |
| Vercel AI Gateway aggregator | fast | $20per Mtok in$100per Mtok out$2per Mtok cached$25per Mtok cache writeai-gateway.vercel.sh · read 2026-09-20 |
| kilo 0 ways | $10per Mtok in$50per Mtok out$1per Mtok cached | |
| kilo aggregatornot on the seller's list today | standard | $10per Mtok in$50per Mtok out$1per Mtok cachedmodels.dev · read 2026-09-05 |
| llmgateway 0 ways | $10per Mtok in$50per Mtok out$1per Mtok cached | |
| llmgateway aggregatornot on the seller's list today | standard | $10per Mtok in$50per Mtok out$1per Mtok cachedmodels.dev · read 2026-09-05 |
| llmgateway-providers 0 ways | $10per Mtok in$50per Mtok out$1per Mtok cached | |
| llmgateway-providers aggregatornot on the seller's list today | standard | $10per Mtok in$50per Mtok out$1per Mtok cachedmodels.dev · read 2026-09-05 |
| merge-gateway 0 ways | $10per Mtok in$50per Mtok out$1per Mtok cached | |
| merge-gateway aggregatornot on the seller's list today | standard | $10per Mtok in$50per Mtok out$1per Mtok cachedmodels.dev · read 2026-09-05 |
| nano-gpt 0 ways | $10per Mtok in$50per Mtok out$1per Mtok cached | |
| nano-gpt aggregatornot on the seller's list today | standard | $10per Mtok in$50per Mtok out$1per Mtok cachedmodels.dev · read 2026-09-05 |
Measured 47 standings
| Place | Board | Metric | Score |
|---|---|---|---|
| 1stof 132 | Ale bench — Epoch AI | Performance (max) | 2951.32 |
| 1stof 290 | Arc agi 2 — Epoch AI | Score (max) | 95 |
| 1stof 41 | Balrog — Epoch AI | Average progress (max) | 0.683 |
| 1stof 69 | Deepswe — Epoch AI | Pass@1 (xhigh) | 0.741 |
| 1stof 21 | Frontierswe — Epoch AI | Score (max) | 65.508 |
| 1stof 56 | Humanity's Last Exam (Epoch AI replication) | Accuracy | 54.8 |
| 1stof 142 | Mystery game puzzles — Epoch AI | mean_score (max) | 84 |
| 1stof 25 | Weirdml v3 — Epoch AI | Score (xhigh) | 42.22 |
| 2ndof 298 | ARC-AGI-2 (Semi-Private) | Score (%) (max) | 95 |
| 2ndof 154 | Chess puzzles — Epoch AI | mean_score (max) | 74 |
| 2ndof 28 | Ebr bench — Epoch AI | mean_score (max) | 76.19 |
| 2ndof 14 | Frontiermath erdos — Epoch AI | mean_score (max) | 2.941 |
| 2ndof 78 | Frontiermath tier 4 v2 — Epoch AI | mean_score (high) | 97.6 |
| 2ndof 122 | Frontiermath tiers 1 3 v2 — Epoch AI | mean_score (max) | 93.684 |
| 2ndof 326 | GPQA Diamond | mean_score (max) | 95.77 |
| 2ndof 56 | Gdp pdf — Epoch AI | GDP.pdf score (max) | 0.342 |
| 2ndof 42 | Gso — Epoch AI | Score OPT@1 | 0.794 |
| 2ndof 141 | LMArena · WebDev | Score (Elo) (max) | 1788 |
| 2ndof 18 | Rli — Epoch AI | Score | 20.83 |
| 2ndof 93 | Simpleqa verified — Epoch AI | mean_score (max) | 75.6 |
| 2ndof 71 | Vending bench 2 — Epoch AI | Score | 15514.702 |
| 2ndof 135 | Webdev arena — Epoch AI | Arena Score (max) | 1800.28 |
| 2ndof 173 | Weirdml — Epoch AI | Accuracy (max) | 93.28 |
| 3rdof 298 | ARC-AGI-1 (Semi-Private) | Score (%) (xhigh) | 98.5 |
| 3rdof 309 | Arc agi — Epoch AI | Score (xhigh) | 98.5 |
| 3rdof 260 | Critpt — Epoch AI | Accuracy (max) | 31.714 |
| 3rdof 35 | Furniture assembly — Epoch AI | mean_score (max) | 80 |
| 4thof 282 | ARC-AGI-2 (Public Eval) | Score (%) (xhigh) | 97.92 |
| 4thof 86 | Artificial Analysis · Intelligence Index | Artificial Analysis Intelligence Index (max) | 57 |
| 4thof 35 | Blueprint bench 2 — Epoch AI | Score | 49.675 |
| 4thof 46 | Frontiercode — Epoch AI | Main score (max) | 0.533 |
| 4thof 53 | LMArena · Agent | Net Improvement (%) | 13.09 |
| 5thof 84 | Proofbench — Epoch AI | Accuracy | 99 |
| 6thof 273 | ARC-AGI-1 (Public Eval) | Score (%) (xhigh) | 99 |
| 7thof 163 | Artificial Analysis · Coding Index | Index | 76.9 |
| 9thof 306 | OTIS Mock AIME 2024-2025 | mean_score (max) | 100 |
| 10thof 131 | Artificial Analysis · Agentic Index | Index | 51 |
| 10thof 235 | Dtbench — Epoch AI | Accuracy (xhigh) | 97.33 |
| 10thof 30 | Mirrorcode — Epoch AI | mean_score (max) | 60 |
| 11thof 60 | Apex agents — Epoch AI | Pass@1 score | 0.647 |
| 11thof 15 | Btf3 — Epoch AI | Pooled score (low) | 0.135 |
| 16thof 112 | ARC-AGI-3 (Semi-Private) | Score (%) (max) | 62.71 |
| 16thof 197 | Lmca — Epoch AI | Score (xhigh) | 64.37 |
| 19thof 165 | LMArena · Vision | Score (Elo) (max) | 1286 |
| 35thof 414 | LMArena · Text | Score (Elo) (max) | 1475 |
| 41stof 235 | Scicode — Epoch AI | Score (max) | 56.481 |
| 46thof 111 | Vectara Hallucination Leaderboard | Hallucination Rate | 8.7 |
About
GPT 6 Astra — a text model from OpenAI, sold by 14 companies from $10 in and $50 out per million tokens, placed 1st of 290 on Arc agi 2 — Epoch AI.
It takes text, images and files and returns text, with a context window of 272,000 tokens. It was published in September 2026, trained on material up to April 2026. Its sellers say it can reason step by step and call a tool. The catalogue files it under reasoning, code and agents. Fourteen companies sell it. The cheapest is $10 in and $50 out per million tokens at Databricks. Beside the standard rate there are faster, batch, regional, separately routed and flex lanes. It has been measured on 47 boards, and stands best at 1st of 290 on Arc agi 2 — Epoch AI.
Every current figure
- Maker
- OpenAI
- Register
- model
- Takes
- text + image + file
- Returns
- text
- Context
- 272,000 tokens
- Published
- September 2026
- Knowledge to
- April 2026
- Licence
- not read
- Sellers
- 14
- Price
- $10 in and $50 out per million tokens — Databricks
- Boards
- 47
- Best place
- 1st of 290 — Arc agi 2 — Epoch AI
Known as 1 name
openai/gpt-6-astra