In short
How they differ.
Worked out from the figures below. Prices and specs are the makers' own, read Oct 20, 2026.
- Claude Opus 5.5 is cheaper per output token: $20 against $50 per million tokens, so GPT-6 Astra costs 2.5 times as much.
- GPT-6 Astra reads more at once: a context window of 1M tokens against 1M.
- On the 11 independent boards that rank both, Claude Opus 5.5 places higher on 9 and GPT-6 Astra on 2.
Side by side
The specs.
| Claude Opus 5.5 | GPT-6 Astra | |
|---|---|---|
| Made by | Anthropic · US | OpenAI · US |
| Released | Sep 22, 2026 | Sep 3, 2026 |
| Price per 1M tokens | $4 in / $20 out | $10 in / $50 out |
| Context window | 1M | 1M |
| Takes in | text, image | text, image |
| Download and run it yourself | No | No |
| Size | Not published | Not published |
| Maker says | “For long-running agentic coding and knowledge work” | “Our most capable model for the most demanding work.” |
| Sources | Prices, Oct 20, 2026 · Description | Prices, Oct 20, 2026 · Description |
Independent tests
Where they place.
Position on each board we follow (1 is best). Epoch AI's and Vectara's scores are shown; the other boards are linked for their own figures.
| Board | Claude Opus 5.5 | GPT-6 Astra |
|---|---|---|
| Epoch Capabilities IndexEpoch AI, Oct 20, 2026 | #1166.82 | #2166.41 |
| LMArena textLMArena, Oct 15, 2026 | #3 | Not listed |
| LMArena creative writingLMArena, Oct 15, 2026 | #1 | Not listed |
| LMArena coding chatsLMArena, Oct 15, 2026 | #6 | #9 |
| LMArena WebDevLMArena, Oct 19, 2026 | #1 | #2 |
| LMArena mathLMArena, Oct 15, 2026 | #14 | Not listed |
| LMArena visionLMArena, Oct 19, 2026 | #10 | Not listed |
| LMArena documentsLMArena, Sep 13, 2026 | Not listed | #16 |
| LMArena Agent ArenaLMArena, Oct 19, 2026 | #1 | #2 |
| Humanity's Last ExamScale AI | #2 | #1 |
| SWE-Bench ProScale AI, Sep 22, 2026 | Not listed | #5 |
| Remote Labor IndexScale AI | #1 | #2 |
| Vals IndexVals AI, Oct 19, 2026 | #2 | #5 |
| Terminal-Bench 4.0Vals AI, Oct 19, 2026 | #1 | #3 |
| Vibe Code BenchVals AI, Oct 18, 2026 | #5 | #7 |
| ProofBenchVals AI, Oct 12, 2026 | #2 | #6 |
| CUA-bench (games)Vals AI, Oct 13, 2026 | #2 | #1 |
| Vectara hallucination testVectara, Oct 16, 2026 | Not listed | #458.7% |
More comparisons