Model comparison
Claude Sonnet 4.5 vs Grok 4
Grok 4 is the stronger model overall, scoring 48.1 to 44.1 on the Noometry Index.
Last verified . 41 shared benchmarks.
Summary
- They share 41 benchmarks with published results for both. Claude Sonnet 4.5 scores higher in 4 categories and Grok 4 in 6 categories; 10 gaps are clear of the uncertainty.
- The widest gap is in long context, where Grok 4 leads 63.1 to 45.2.
- The biggest single-benchmark swing is GDPval: 42.5% for Claude Sonnet 4.5 and 21.1% for Grok 4.
Side by side
| Claude Sonnet 4.5 | Grok 4 | |
|---|---|---|
| Provider | Anthropic | xAI |
| Noometry Index | 44.1 | 48.1 |
| Released | 2025-09-29 | 2025-07-09 |
| Weights | Proprietary | Proprietary |
| Context window | 200K | — |
| Max output | 64K | — |
| Input $ / M tokens | $3 | — |
| Output $ / M tokens | $15 | — |
| Results tracked | 73 | 48 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding Grok 4 leads
Claude Sonnet 4.5: 47.3 (#61), Grok 4: 50.3 (#46)
| Benchmark | Claude Sonnet 4.5 | Grok 4 |
|---|---|---|
| WeirdML | 47.7% | 45.7% |
| LMArena Coding | 1489 | 1408 |
| SWE-bench Verified | 71.3% | — |
| SWE-bench Verified (bash only) | 71.4% | — |
| Aider Polyglot | — | 79.6% |
| LMArena WebDev | 1393 | — |
| SWE-bench Multilingual | 67% | — |
| SciCode | 44.7% | — |
| GSO | 14.7% | — |
| ALE-Bench | 796.15 | — |
| AlgoTune | 1.52 | — |
Agentic & Tool Use Claude Sonnet 4.5 leads
Claude Sonnet 4.5: 38.3 (#32), Grok 4: 32.3 (#68)
| Benchmark | Claude Sonnet 4.5 | Grok 4 |
|---|---|---|
| Terminal-Bench | 46.5% | 27.2% |
| Berkeley Function Calling Leaderboard | 73.2% | 63% |
| GDPval | 42.5% | 21.1% |
| Cybench | 60% | 43% |
| DeepResearch Bench | 52.6% | 47.3% |
| LMArena Search | 1159 | 1142 |
| METR Time Horizons | 67.4% | 66.6% |
| Remote Labor Index | 2.1% | — |
| τ²-bench Airline | 72% | — |
| τ²-bench Banking | 25.3% | — |
| τ²-bench Retail | 72.4% | — |
| τ²-bench Telecom | 84.9% | — |
| OSWorld | 62.9% | — |
| BALROG | — | 43.6% |
| Vending-Bench 2 | 3,839 | — |
Reasoning Grok 4 leads
Claude Sonnet 4.5: 26.9 (#125), Grok 4: 36.7 (#65)
| Benchmark | Claude Sonnet 4.5 | Grok 4 |
|---|---|---|
| ARC-AGI-2 | 13.6% | 16% |
| SimpleBench | 54.3% | 60.5% |
| Kagi LLM Benchmark | 57.9% | 73.6% |
| ARC-AGI-1 | 63.7% | 66.7% |
| Chess Puzzles | 12% | 28% |
| LMArena Hard Prompts | 1462 | 1409 |
| Epoch Capabilities Index | 146.84 | 146.44 |
| ForecastBench | 61.9 | 60.9 |
| NYT Connections (extended) | 37.3% | — |
| CritPt | 1.1% | — |
| EnigmaEval | 6% | — |
| EBR-Bench | 2.4% | — |
| Mystery Game Puzzles | 17% | — |
| DTBench | 83.2% | — |
| LMCA | 38.8% | — |
Math Grok 4 leads
Claude Sonnet 4.5: 32.3 (#216), Grok 4: 48.4 (#64)
| Benchmark | Claude Sonnet 4.5 | Grok 4 |
|---|---|---|
| OTIS Mock AIME 2024-2025 | 77.8% | 84% |
| Omni-MATH | 55.3% | 60.3% |
| LMArena Math | 1449 | 1422 |
| FrontierMath (Feb 2025 set) | 15.2% | 19.7% |
| FrontierMath Tier 4 (v1) | 4.2% | 2.1% |
| FrontierMath (Tiers 1-3) | 23.9% | — |
| FrontierMath Tier 4 | 2.4% | — |
| ProofBench | 19% | — |
| MATH Level 5 | 97.7% | — |
Knowledge Grok 4 leads
Claude Sonnet 4.5: 48.4 (#76), Grok 4: 53.8 (#55)
| Benchmark | Claude Sonnet 4.5 | Grok 4 |
|---|---|---|
| GPQA Diamond | 82.3% | 87% |
| MMLU-Pro | 86.9% | 85.1% |
| GPQA (HELM) | 68.6% | 72.7% |
| LMArena Expert | 1482 | 1415 |
| Humanity's Last Exam | 13.7% | — |
| SimpleQA Verified | 30.7% | — |
| Confabulations | — | 12.4% |
| Vectara Hallucination Rate | 12% | — |
Multimodal Claude Sonnet 4.5 leads
Claude Sonnet 4.5: 34.8 (#89), Grok 4: 33.7 (#94)
| Benchmark | Claude Sonnet 4.5 | Grok 4 |
|---|---|---|
| LMArena Vision | — | 1210 |
| GeoBench | — | 45% |
| VPCT | 39.8% | — |
| LMArena Document | 1450 | — |
Multilingual Claude Sonnet 4.5 leads
Claude Sonnet 4.5: 53.4 (#69), Grok 4: 51.8 (#103)
| Benchmark | Claude Sonnet 4.5 | Grok 4 |
|---|---|---|
| LMArena Non-English | 1425 | 1403 |
| LMArena Chinese | 1459 | 1427 |
| LMArena French | 1458 | 1418 |
| LMArena German | 1427 | 1429 |
| LMArena Japanese | 1390 | 1394 |
| LMArena Korean | 1403 | 1377 |
| LMArena Russian | 1437 | 1410 |
| LMArena Spanish | 1457 | 1420 |
Instruction Following Grok 4 leads
Claude Sonnet 4.5: 75.0 (#78), Grok 4: 79.2 (#5)
| Benchmark | Claude Sonnet 4.5 | Grok 4 |
|---|---|---|
| IFEval | 85% | 94.9% |
| LMArena Instruction Following | 1459 | 1387 |
Long Context Grok 4 leads
Claude Sonnet 4.5: 45.2 (#46), Grok 4: 63.1 (#4)
| Benchmark | Claude Sonnet 4.5 | Grok 4 |
|---|---|---|
| LMArena Longer Query | 1476 | 1409 |
| Fiction.LiveBench | — | 94.4% |
Writing & Preference Claude Sonnet 4.5 leads
Claude Sonnet 4.5: 66.5 (#34), Grok 4: 58.5 (#116)
| Benchmark | Claude Sonnet 4.5 | Grok 4 |
|---|---|---|
| LMArena Text | 1439 | 1411 |
| LMArena Creative Writing | 1442 | 1397 |
| WildBench | 85.4% | 79.7% |
| LMArena Multi-Turn | 1465 | 1416 |
| Short-Story Creative Writing | — | 76.9% |
| EQ-Bench Creative Writing | 1678 | — |
Frequently asked questions
Is Claude Sonnet 4.5 better than Grok 4?
Grok 4 is the stronger model overall, scoring 48.1 to 44.1 on the Noometry Index.
Is Claude Sonnet 4.5 or Grok 4 better for coding?
Grok 4 scores higher on coding benchmarks: 50.3 versus 47.3 in the Noometry coding category.
How many benchmarks do Claude Sonnet 4.5 and Grok 4 share?
41 benchmarks have published results for both models. Claude Sonnet 4.5 has 73 scored results on Noometry and Grok 4 has 48.