- Context window
- 1.1M
- Input
- Image · Text
- Output
- Text
- Released
- 2026-09-03
- Max output
- 128K
- Knowledge cutoff
- 2026-04-30
- Weights
- Closed
- Capabilities
- Attachments, Reasoning, Structured, Tool call
IN
$10.00 – $20.00
USD / 1,000,000 token
$10.00 · prompt: <=272000 input tokens
$20.00 · prompt: >272000 input tokens
Cache read
$1.00 – $2.00
USD / 1,000,000 token
$1.00 · prompt: <=272000 input tokens
$2.00 · prompt: >272000 input tokens
Cache write
$12.50 – $50.00
USD / 1,000,000 token
$12.50 · prompt: <=272000 input tokens; cache minimum lifetime 30m; total write rate (not surcharge)
$25.00 · prompt: >272000 input tokens; cache minimum lifetime 30m; total write rate (not surcharge)
$25.00 · prompt: <=272000 input tokens; cache minimum lifetime 12h; total write rate (not surcharge)
$50.00 · prompt: >272000 input tokens; cache minimum lifetime 12h; total write rate (not surcharge)
OUT
$50.00 – $75.00
USD / 1,000,000 token
$50.00 · prompt: <=272000 input tokens
$75.00 · prompt: >272000 input tokens
Performance
Overall capability
72.9/100
- Coding
- 72.2
- Reasoning
- 80.4
- Professional work
- 60.5
- Knowledge & language
- 72.3
| Arena Agent Overall | Professional work | LMArena | 2026-11-08 | |
| Code Migration | Coding | Vals AI | 2026-11-05 | |
| APEX-Agents 1.1 | Professional work | Mercor | 2026-11-05 | |
| AutomationBench-AAGPT-6 Astra (Max) | Professional work | Artificial Analysis | 2026-11-05 | |
| Terminal-Bench Science · AAGPT-6 Astra (Max) | Reasoning | Artificial Analysis | 2026-11-05 | |
| CritPtGPT-6 Astra (Max) | Reasoning | Artificial Analysis | 2026-11-05 | |
| Vibe Code Bench 1–100 | Coding | Vals AI | 2026-11-05 | |
| τ³-Banking | Professional work | Mercor | 2026-11-05 |
Providers
| Input | Output | Cache read | Cache write | ||||
|---|---|---|---|---|---|---|---|
| OpenAI | gpt-6-astra | Openai responses | global | ||||
| OpenRouter | openai/gpt-6-astra | Openai chat | global |
ModelRegistry · Catalog updated 2026-11-10