Model comparison
Claude Sonnet 4 vs Grok-3 mini
Claude Sonnet 4 and Grok-3 mini score almost the same on the Noometry Index (40.8 vs 41.2), so choose on price, context window or the category you care about most.
Last verified . 35 shared benchmarks.
Summary
- They share 35 benchmarks with published results for both. Claude Sonnet 4 scores higher in 4 categories and Grok-3 mini in 4 categories; 8 gaps are clear of the uncertainty.
- The widest gap is in reasoning, where Claude Sonnet 4 leads 22.9 to 13.6.
- The biggest single-benchmark swing is Omni-MATH: 60.2% for Claude Sonnet 4 and 31.8% for Grok-3 mini.
Side by side
| Claude Sonnet 4 | Grok-3 mini | |
|---|---|---|
| Provider | Anthropic | xAI |
| Noometry Index | 40.8 | 41.2 |
| Released | 2025-05-22 | 2025-04-09 |
| Weights | Proprietary | Proprietary |
| Context window | 200K | — |
| Max output | 64K | — |
| Input $ / M tokens | $3 | — |
| Output $ / M tokens | $15 | — |
| Results tracked | 58 | 35 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding Claude Sonnet 4 leads
Claude Sonnet 4: 43.5 (#88), Grok-3 mini: 40.8 (#131)
| Benchmark | Claude Sonnet 4 | Grok-3 mini |
|---|---|---|
| Aider Polyglot | 61.3% | 49.3% |
| WeirdML | 46.1% | 42.6% |
| LMArena Coding | 1414 | 1379 |
| SWE-bench Verified (bash only) | 64.9% | — |
| SciCode | 40% | — |
| GSO | 4.9% | — |
| ALE-Bench | 655.35 | — |
Agentic & Tool Use Not comparable
Claude Sonnet 4: 38.5 (#31), Grok-3 mini: —
| Benchmark | Claude Sonnet 4 | Grok-3 mini |
|---|---|---|
| TheAgentCompany | 33.1% | — |
| Cybench | 35% | — |
| DeepResearch Bench | 46.6% | — |
| OSWorld | 43.9% | — |
| METR Time Horizons | 62% | — |
Reasoning Claude Sonnet 4 leads
Claude Sonnet 4: 22.9 (#187), Grok-3 mini: 13.6 (#334)
| Benchmark | Claude Sonnet 4 | Grok-3 mini |
|---|---|---|
| ARC-AGI-2 | 5.9% | 0.4% |
| Kagi LLM Benchmark | 73% | 61.3% |
| ARC-AGI-1 | 40% | 16.5% |
| LMArena Hard Prompts | 1372 | 1375 |
| Epoch Capabilities Index | 141.69 | 140.35 |
| SimpleBench | 45.5% | — |
| CritPt | 0.3% | — |
| EnigmaEval | 3.1% | — |
| DTBench | 77.1% | — |
| LMCA | 29% | — |
| ForecastBench | 60.2 | — |
Math Claude Sonnet 4 leads
Claude Sonnet 4: 43.3 (#80), Grok-3 mini: 42.1 (#85)
| Benchmark | Claude Sonnet 4 | Grok-3 mini |
|---|---|---|
| OTIS Mock AIME 2024-2025 | 71.1% | 77.8% |
| Omni-MATH | 60.2% | 31.8% |
| LMArena Math | 1375 | 1386 |
| MATH Level 5 | 84.4% | 90.9% |
| FrontierMath (Feb 2025 set) | 4.1% | 5.9% |
| FrontierMath Tier 4 (v1) | 0% | — |
Knowledge Grok-3 mini leads
Claude Sonnet 4: 41.8 (#108), Grok-3 mini: 46.4 (#81)
| Benchmark | Claude Sonnet 4 | Grok-3 mini |
|---|---|---|
| GPQA Diamond | 79.2% | 76.3% |
| MMLU-Pro | 84.3% | 79.9% |
| Confabulations | 13.2% | 10.8% |
| GPQA (HELM) | 70.6% | 67.5% |
| LMArena Expert | 1372 | 1395 |
| Humanity's Last Exam | 7.8% | — |
| Vectara Hallucination Rate | 10.3% | — |
Multimodal Not comparable
Claude Sonnet 4: 26.2 (#121), Grok-3 mini: —
| Benchmark | Claude Sonnet 4 | Grok-3 mini |
|---|---|---|
| LMArena Vision | 1191 | — |
| GeoBench | 37% | — |
| VPCT | 34% | — |
| MindCube | 44.8% | — |
Multilingual Grok-3 mini leads
Claude Sonnet 4: 46.7 (#156), Grok-3 mini: 48.1 (#145)
| Benchmark | Claude Sonnet 4 | Grok-3 mini |
|---|---|---|
| LMArena Non-English | 1333 | 1352 |
| LMArena Chinese | 1350 | 1387 |
| LMArena French | 1363 | 1357 |
| LMArena German | 1331 | 1349 |
| LMArena Japanese | 1302 | 1342 |
| LMArena Korean | 1291 | 1335 |
| LMArena Russian | 1355 | 1353 |
| LMArena Spanish | 1357 | 1381 |
Instruction Following Grok-3 mini leads
Claude Sonnet 4: 71.7 (#145), Grok-3 mini: 78.5 (#9)
| Benchmark | Claude Sonnet 4 | Grok-3 mini |
|---|---|---|
| IFEval | 84% | 95.1% |
| LMArena Instruction Following | 1376 | 1357 |
Long Context Grok-3 mini leads
Claude Sonnet 4: 33.7 (#259), Grok-3 mini: 41.0 (#147)
| Benchmark | Claude Sonnet 4 | Grok-3 mini |
|---|---|---|
| Fiction.LiveBench | 46.9% | 66.7% |
| LMArena Longer Query | 1398 | 1372 |
Writing & Preference Claude Sonnet 4 leads
Claude Sonnet 4: 57.1 (#132), Grok-3 mini: 52.5 (#169)
| Benchmark | Claude Sonnet 4 | Grok-3 mini |
|---|---|---|
| LMArena Text | 1351 | 1370 |
| LMArena Creative Writing | 1345 | 1342 |
| Short-Story Creative Writing | 81.4% | 73.5% |
| WildBench | 83.8% | 65.1% |
| LMArena Multi-Turn | 1376 | 1355 |
| EQ-Bench Creative Writing | 1483 | — |
Frequently asked questions
Is Claude Sonnet 4 better than Grok-3 mini?
Claude Sonnet 4 and Grok-3 mini score almost the same on the Noometry Index (40.8 vs 41.2), so choose on price, context window or the category you care about most.
Is Claude Sonnet 4 or Grok-3 mini better for coding?
Claude Sonnet 4 scores higher on coding benchmarks: 43.5 versus 40.8 in the Noometry coding category.
How many benchmarks do Claude Sonnet 4 and Grok-3 mini share?
35 benchmarks have published results for both models. Claude Sonnet 4 has 58 scored results on Noometry and Grok-3 mini has 35.