Claude Opus 4.6 vs Grok 4.6
Claude Opus 4.6 
Grok 4.6 Verdict
- Claude Opus 4.6 and Grok 4.6 (medium) are level on quality (63.1 vs 63.8).
- Claude Opus 4.6 is stronger in agents & tools, human preference, instruction following, maths, multimodal, reasoning.
- Grok 4.6 (medium) is stronger in coding, composite, knowledge, long context.
- Grok 4.6 (medium) is 3.3× cheaper ($3.00 vs $10.00 per 1M blended).
- Grok 4.6 (medium) streams 1.4× faster (55 vs 38 tokens per second).
| Metric | Claude Opus 4.6 | Grok 4.6 (medium) |
|---|---|---|
| BenchLeader Index | 63.1 | 63.8 |
| Agents & tools score | 68.8 | – |
| Coding score | 62.3 | 63.9 |
| Composite score | 69.2 | 83.2 |
| Human preference score | 64.6 | – |
| Instruction following score | 53.7 | – |
| Knowledge score | 70.5 | 77.3 |
| Long context score | 65.5 | 67.1 |
| Maths score | 65.6 | – |
| Multimodal score | 63.6 | – |
| Reasoning score | 62.3 | 62.1 |
| Blended price $/M | $10.00 | $3.00 |
| Output speed | 38 tok/s | 55 tok/s |
| Time to first answer | 2.0 s | 36.7 s |
| Context window | 1M | 500k |
| GPQA Diamond | 90.5% | – |
| OTIS Mock AIME | 94.4% | – |
| SWE-bench Verified (Epoch) | 78.7% | – |
| Humanity's Last Exam | 19.0% | – |
| Terminal-Bench | 79.8% | – |
| SimpleBench | 67.6% | – |
| SciCode | – | 54.6% |
| Cybench | 93.0% | – |
| Remote Labor Index | 4.2% | – |
| WeirdML | 77.9% | – |
| APEX-Agents | 32.4% | – |
| FrontierCode | 26.6% | – |
| GSO-Bench | 33.3% | – |
| Epoch Capabilities Index | 155.3 | – |
| LMArena Text | 1498 | – |
| LMArena Hard Prompts | 1527 | – |
| LMArena Coding | 1546 | – |
| LMArena WebDev | 1537 | – |
| LMArena Vision | 1311 | – |
| AA Intelligence Index | 31.9 | 43.0 |
| IFBench | 53.1% | – |
| AA-LCR | 78.0% | 81.0% |
| MMMU-Pro | 75.4% | – |
| AA-Omniscience | 13.7 | 28 |
| Terminal-Bench Hard | 48.5% | – |
| GPQA Diamond (AA) | 89.6% | 93.5% |
| Humanity's Last Exam (AA) | 39.9% | 42.1% |
| SciCode (AA) | – | 55.9% |
| τ²-Bench Telecom (AA) | 92.1% | – |
| HiL-Bench | 38.3% | – |
| EQ-Bench 4 | 1223 | – |
| Kagi LLM Benchmark | 72.4% | – |
| SWE-bench Verified (bash only) | 75.6% | – |
| SWE-bench Verified (any scaffold) | 75.6% | – |
| ARC-AGI-1 | – | 87.5% |
| ARC-AGI-2 | – | 61.3% |
Data as of 2026-09-09. Best configuration of each model; every score links to its source on the model pages.