Claude Opus 5 vs GPT-6 Astra
Claude Opus 5 
GPT-6 Astra Verdict
- Claude Opus 5 and GPT-6 Astra (max) are level on quality (70.0 vs 70.6).
- Claude Opus 5 is stronger in agents & tools, composite, human preference, long context, multimodal, reasoning.
- GPT-6 Astra (max) is stronger in coding, knowledge, maths.
- Claude Opus 5 is 2.2× cheaper ($10.00 vs $22.00 per 1M blended).
| Metric | Claude Opus 5 | GPT-6 Astra (max) |
|---|---|---|
| BenchLeader Index | 70.0 | 70.6 |
| Agents & tools score | 71.6 | 68.2 |
| Coding score | 74.4 | 79.0 |
| Composite score | 92.9 | 73.1 |
| Human preference score | 78.2 | – |
| Knowledge score | 70.5 | 79.8 |
| Long context score | 66.2 | – |
| Maths score | 67.1 | 77.0 |
| Multimodal score | 69.4 | – |
| Reasoning score | 71.1 | 70.9 |
| Blended price $/M | $10.00 | $22.00 |
| Output speed | 52 tok/s | 54 tok/s |
| Time to first answer | 81.6 s | 328.6 s |
| Context window | 1M | 1.1M |
| GPQA Diamond | 92.9% | 95.8% |
| FrontierMath Tiers 1–3 | – | 93.7% |
| FrontierMath Tier 4 | – | 97.6% |
| OTIS Mock AIME | 97.8% | 100.0% |
| SimpleQA Verified | – | 75.6% |
| Terminal-Bench | – | 58.2% |
| SimpleBench | 80.6% | – |
| SciCode | – | 56.5% |
| WeirdML | 86.3% | – |
| FrontierCode | – | 53.3% |
| Epoch Capabilities Index | 162.6 | – |
| LMArena WebDev | – | 1796 |
| LMArena Agent | – | 12.5 |
| LiveBench | – | 82.2% |
| LiveBench Reasoning | – | 92.7% |
| LiveBench Coding | – | 80.4% |
| LiveBench Agentic Coding | – | 57.3% |
| LiveBench Mathematics | – | 96.8% |
| LiveBench Data Analysis | – | 83.0% |
| LiveBench Language | – | 89.4% |
| AA Intelligence Index | 50.7 | – |
| AA-LCR | 79.3% | – |
| MMMU-Pro | 84.7% | – |
| AA-Omniscience | 37.1 | – |
| GPQA Diamond (AA) | 93.2% | – |
| Humanity's Last Exam (AA) | 54.9% | – |
| SciCode (AA) | 56.4% | – |
| LiveCodeBench | 89.0% | – |
| MMLU-Pro | 91.6% | – |
| IOI | 91.7% | – |
| LegalBench | 87.0% | – |
| CorpFin | 73.2% | – |
| TaxEval | 75.1% | – |
| Terminal-Bench 2.1 (Vals) | 84.6% | 87.3% |
| SWE-bench (Vals) | 97.0% | – |
| GPQA Diamond (Vals) | 93.4% | – |
| Vals Index | 67.2 | 66.6 |
| HiL-Bench | 57.0% | – |
| EQ-Bench 4 | 1385 | – |
| ARC-AGI-1 | – | 97.5% |
| ARC-AGI-2 | – | 95.0% |
| ARC-AGI-3 | – | 62.7% |
Data as of 2026-09-09. Best configuration of each model; every score links to its source on the model pages.