Claude Opus 4.7 vs Grok 4.5
Claude Opus 4.7 
Grok 4.5 Verdict
- Claude Opus 4.7 leads on quality: 65.5 vs 63.4.
- Claude Opus 4.7 is stronger in agents & tools, coding, composite, human preference, instruction following, maths, multimodal.
- Grok 4.5 is stronger in knowledge, long context, reasoning.
- Grok 4.5 is 3.3× cheaper ($3.00 vs $10.00 per 1M blended).
| Metric | Claude Opus 4.7 | Grok 4.5 |
|---|---|---|
| BenchLeader Index | 65.5 | 63.4 |
| Agents & tools score | 69.4 | 58.6 |
| Coding score | 64.8 | 59.9 |
| Composite score | 80.3 | 66.0 |
| Human preference score | 68.8 | 66.9 |
| Instruction following score | 58.5 | – |
| Knowledge score | 66.4 | 76.0 |
| Long context score | 65.9 | 66.2 |
| Maths score | 65.2 | – |
| Multimodal score | 65.7 | 64.8 |
| Reasoning score | 66.3 | 68.9 |
| Blended price $/M | $10.00 | $3.00 |
| Output speed | 46 tok/s | 56 tok/s |
| Time to first answer | 21.2 s | 12.7 s |
| Context window | 1M | 500k |
| Humanity's Last Exam | 36.2% | – |
| Terminal-Bench | 80.2% | – |
| SimpleBench | 61.7% | 70.0% |
| WeirdML | 76.4% | 46.4% |
| APEX-Agents | – | 34.2% |
| FrontierCode | 38.5% | 42.4% |
| GSO-Bench | 44.1% | – |
| Epoch Capabilities Index | 156.3 | 153.9 |
| LMArena Text | 1495 | 1471 |
| LMArena Hard Prompts | 1519 | 1495 |
| LMArena Coding | 1547 | 1523 |
| LMArena WebDev | 1557 | 1556 |
| LMArena Vision | 1317 | 1291 |
| LMArena Agent | – | 3.9 |
| LiveBench | – | 75.8% |
| LiveBench Reasoning | – | 87.2% |
| LiveBench Coding | – | 68.6% |
| LiveBench Agentic Coding | – | 56.5% |
| LiveBench Mathematics | – | 90.8% |
| LiveBench Data Analysis | – | 73.0% |
| LiveBench Language | – | 82.8% |
| AA Intelligence Index | 40.7 | 39.1 |
| IFBench | 58.6% | – |
| AA-LCR | 78.7% | 79.3% |
| MMMU-Pro | 78.8% | 80.4% |
| AA-Omniscience | 27.3 | 25.3 |
| Terminal-Bench Hard | 51.5% | – |
| GPQA Diamond (AA) | 91.4% | 93.1% |
| Humanity's Last Exam (AA) | 42.3% | 42.7% |
| SciCode (AA) | – | 55.0% |
| τ²-Bench Telecom (AA) | 88.6% | – |
| AIME (Vals) | 96.3% | – |
| LiveCodeBench | 85.1% | – |
| MMLU-Pro | 89.9% | – |
| IOI | 47.1% | – |
| LegalBench | 85.3% | – |
| CorpFin | 66.1% | – |
| TaxEval | 75.3% | – |
| Terminal-Bench 2.1 (Vals) | 68.5% | – |
| SWE-bench (Vals) | 82.0% | – |
| GPQA Diamond (Vals) | 90.2% | – |
| Vals Index | 56.1 | – |
| HiL-Bench | 41.7% | – |
| EQ-Bench 4 | 1311 | – |
| Kagi LLM Benchmark | 73.3% | 83.5% |
Data as of 2026-09-09. Best configuration of each model; every score links to its source on the model pages.