| Benchmark | Claude Opus 4.7 | Muse Spark |
|---|---|---|
| AA Agentic Index | 64.6 | 28.7 |
| AA Intelligence | 55 | 44.3 |
| AA-LCR | 75.3 | 77 |
| AA-Omniscience | 27.3 | 7.2 |
| ARC-AGI-2 | 75.8 | 42.5 |
| arena_elo | 1494 | 1488 |
| arena_text_factuality | 1474 | 1467 |
| arena_vision | 1304 | 1295 |
| Artificial Analysis Coding Index | 73.6 | 58.6 |
| coding_arena_elo | 1557 | 1540 |
| critpt | 12 | 11.3 |
| cybergym | 73.8 | 43.5 |
| frontiermath_tier_4 | 22.9 | 14.6 |
| gdpval | 58.6 | 32.3 |
| GPQA Diamond | 94.2 | 89.5 |
| HLE | 54.7 | 58.4 |
| IFBench | 58.6 | 75.9 |
| itbenchSre | 46.7 | 16.2 |
| MCP Atlas | 79.1 | 82.2 |
| MMMU-Pro | 78.8 | 80.5 |
| OmniScience Accuracy | 48.9 | 49.6 |
| OmniScience Non-Hallucination | 57.7 | 26.8 |
| scicode | 54.5 | 51.5 |
| screenspot_pro_no_tools | 79.5 | 84.1 |
| simpleqa_verified | 50.6 | 66.3 |
| SWE-bench Pro | 64.3 | 55 |
| SWE-bench Verified | 87.6 | 77.4 |
| TauBench V3 - Banking | 34.6 | 19.6 |
| Terminal-Bench 2.0 | 69.4 | 59 |
| Terminal-Bench 2.1 | 83.1 | 62.2 |
| Terminal-Bench Hard | 54.5 | 45.5 |
| τ²-Bench Telecom (AA run) | 88.6 | 91.5 |
| τ³-Bench | 28.9 | 19.6 |
Best tracked score per model per benchmark (default configuration; source-attributed). ↓ marks lower-is-better metrics. Open either model for its full surface, provenance and pricing.