Claude Opus 4.6 vs Grok 4
Claude Opus 4.6 
Grok 4 Verdict
- Claude Opus 4.6 (thinking) leads on quality: 62.5 vs 59.8.
- Claude Opus 4.6 (thinking) is stronger in coding, knowledge, maths, reasoning.
- Grok 4 is stronger in agents & tools, human preference, instruction following, long context, multimodal.
- Grok 4 is 6.4× cheaper ($1.56 vs $10.00 per 1M blended).
| Metric | Claude Opus 4.6 (thinking) | Grok 4 |
|---|---|---|
| BenchLeader Index | 62.5 | 59.8 |
| Coding score | 63.6 | 61.2 |
| Knowledge score | 62.2 | 56.0 |
| Maths score | 65.1 | 58.0 |
| Reasoning score | 75.6 | 58.3 |
| Agents & tools score | – | 53.4 |
| Human preference score | – | 59.8 |
| Instruction following score | – | 67.7 |
| Long context score | – | 77.7 |
| Multimodal score | – | 54.5 |
| Blended price $/M | $10.00 | $1.56 |
| Output speed | 38 tok/s | – |
| Time to first answer | 2.0 s | – |
| Context window | 1M | 256k |
| GPQA Diamond | – | 87.0% |
| OTIS Mock AIME | – | 84.0% |
| Terminal-Bench | – | 27.2% |
| SimpleBench | – | 60.5% |
| Fiction.LiveBench 120k | – | 96.9% |
| Cybench | – | 43.0% |
| WeirdML | – | 45.7% |
| APEX-Agents | – | 15.2% |
| Epoch Capabilities Index | – | 146.4 |
| LMArena Text | – | 1411 |
| LMArena Hard Prompts | – | 1420 |
| LMArena Coding | – | 1435 |
| LMArena Vision | – | 1210 |
| AIME (Vals) | 95.6% | 90.6% |
| LiveCodeBench | 84.7% | 83.3% |
| MMLU-Pro | 89.1% | 85.3% |
| LegalBench | 85.3% | 83.2% |
| CorpFin | 67.0% | 66.0% |
| TaxEval | 76.0% | 65.1% |
| MedQA | 95.4% | 92.5% |
| MGSM | – | 90.9% |
| SWE-bench (Vals) | 78.2% | 57.8% |
| GPQA Diamond (Vals) | 89.7% | 88.1% |
| IMO 2025 | – | 21.4% |
| MathArena Apex | – | 2.1% |
| SWE-Bench Pro | 51.9% | – |
| Kagi LLM Benchmark | 83.6% | 73.6% |
| IFEval (HELM) | – | 94.9% |
| Omni-MATH (HELM) | – | 60.3% |
| WildBench (HELM) | – | 79.7% |
| MMLU-Pro (HELM) | – | 85.1% |
| GPQA Diamond (HELM) | – | 72.6% |
| HELM Capabilities mean | – | 78.5% |
| Aider Polyglot | – | 79.6% |
| ARC-AGI-1 | – | 79.6% |
| ARC-AGI-2 | – | 29.4% |
| BFCL Overall | – | 63.0% |
Data as of 2026-09-13. Best configuration of each model; every score links to its source on the model pages.
Claude Opus 4.6 vs Grok 4: questions
- Is Claude Opus 4.6 better than Grok 4?
- Claude Opus 4.6 (thinking) leads on quality: 62.5 vs 59.8. The BenchLeader Index combines every independent quality benchmark; Claude Opus 4.6 (thinking) is ahead overall as of 2026-09-13, but check the category scores for your use.
- Is Claude Opus 4.6 better than Grok 4 for coding?
- Claude Opus 4.6 scores higher in coding (64 vs 61 on the category index, where 50 is average).
- Which is cheaper, Claude Opus 4.6 or Grok 4?
- Grok 4 is cheaper: $1.56 against $10.00 per million tokens, blended at three input tokens per output token.
- Which has the larger context window?
- Claude Opus 4.6 accepts more context: 1M against 256k tokens.