Model comparison
Claude Sonnet 4 vs Kimi K2 Thinking Turbo
Kimi K2 Thinking Turbo is the stronger model overall, scoring 45.8 to 40.8 on the Noometry Index.
Last verified . 19 shared benchmarks.
Summary
- They share 19 benchmarks with published results for both. Claude Sonnet 4 scores higher in 1 category and Kimi K2 Thinking Turbo in 7 categories; 8 gaps are clear of the uncertainty.
- The widest gap is in long context, where Kimi K2 Thinking Turbo leads 43.2 to 33.7.
- The biggest single-benchmark swing is OTIS Mock AIME 2024-2025: 71.1% for Claude Sonnet 4 and 83.1% for Kimi K2 Thinking Turbo.
- Kimi K2 Thinking Turbo has downloadable open weights; the other is API-only.
Side by side
| Claude Sonnet 4 | Kimi K2 Thinking Turbo | |
|---|---|---|
| Provider | Anthropic | Moonshot AI |
| Noometry Index | 40.8 | 45.8 |
| Released | 2025-05-22 | 2025-11-06 |
| Weights | Proprietary | Open |
| Context window | 200K | — |
| Max output | 64K | — |
| Input $ / M tokens | $3 | — |
| Output $ / M tokens | $15 | — |
| Results tracked | 58 | 21 |
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), Kimi K2 Thinking Turbo: 38.4 (#178)
| Benchmark | Claude Sonnet 4 | Kimi K2 Thinking Turbo |
|---|---|---|
| LMArena Coding | 1414 | 1454 |
| SWE-bench Verified (bash only) | 64.9% | — |
| Aider Polyglot | 61.3% | — |
| LMArena WebDev | — | 1322 |
| SciCode | 40% | — |
| GSO | 4.9% | — |
| WeirdML | 46.1% | — |
| ALE-Bench | 655.35 | — |
Agentic & Tool Use Not comparable
Claude Sonnet 4: 38.5 (#31), Kimi K2 Thinking Turbo: —
| Benchmark | Claude Sonnet 4 | Kimi K2 Thinking Turbo |
|---|---|---|
| TheAgentCompany | 33.1% | — |
| Cybench | 35% | — |
| DeepResearch Bench | 46.6% | — |
| OSWorld | 43.9% | — |
| METR Time Horizons | 62% | — |
Reasoning Kimi K2 Thinking Turbo leads
Claude Sonnet 4: 22.9 (#187), Kimi K2 Thinking Turbo: 32.5 (#75)
| Benchmark | Claude Sonnet 4 | Kimi K2 Thinking Turbo |
|---|---|---|
| LMArena Hard Prompts | 1372 | 1428 |
| ARC-AGI-2 | 5.9% | — |
| SimpleBench | 45.5% | — |
| Kagi LLM Benchmark | 73% | — |
| ARC-AGI-1 | 40% | — |
| CritPt | 0.3% | — |
| Chess Puzzles | — | 20% |
| EnigmaEval | 3.1% | — |
| DTBench | 77.1% | — |
| LMCA | 29% | — |
| Epoch Capabilities Index | 141.69 | — |
| ForecastBench | 60.2 | — |
Math Kimi K2 Thinking Turbo leads
Claude Sonnet 4: 43.3 (#80), Kimi K2 Thinking Turbo: 47.4 (#68)
| Benchmark | Claude Sonnet 4 | Kimi K2 Thinking Turbo |
|---|---|---|
| OTIS Mock AIME 2024-2025 | 71.1% | 83.1% |
| LMArena Math | 1375 | 1429 |
| Omni-MATH | 60.2% | — |
| MATH Level 5 | 84.4% | — |
| FrontierMath (Feb 2025 set) | 4.1% | — |
| FrontierMath Tier 4 (v1) | 0% | — |
Knowledge Kimi K2 Thinking Turbo leads
Claude Sonnet 4: 41.8 (#108), Kimi K2 Thinking Turbo: 50.9 (#69)
| Benchmark | Claude Sonnet 4 | Kimi K2 Thinking Turbo |
|---|---|---|
| GPQA Diamond | 79.2% | 84.2% |
| LMArena Expert | 1372 | 1439 |
| Humanity's Last Exam | 7.8% | — |
| MMLU-Pro | 84.3% | — |
| Confabulations | 13.2% | — |
| Vectara Hallucination Rate | 10.3% | — |
| GPQA (HELM) | 70.6% | — |
Multimodal Not comparable
Claude Sonnet 4: 26.2 (#121), Kimi K2 Thinking Turbo: —
| Benchmark | Claude Sonnet 4 | Kimi K2 Thinking Turbo |
|---|---|---|
| LMArena Vision | 1191 | — |
| GeoBench | 37% | — |
| VPCT | 34% | — |
| MindCube | 44.8% | — |
Multilingual Kimi K2 Thinking Turbo leads
Claude Sonnet 4: 46.7 (#156), Kimi K2 Thinking Turbo: 51.4 (#109)
| Benchmark | Claude Sonnet 4 | Kimi K2 Thinking Turbo |
|---|---|---|
| LMArena Non-English | 1333 | 1398 |
| LMArena Chinese | 1350 | 1456 |
| LMArena French | 1363 | 1425 |
| LMArena German | 1331 | 1390 |
| LMArena Japanese | 1302 | 1357 |
| LMArena Korean | 1291 | 1330 |
| LMArena Russian | 1355 | 1391 |
| LMArena Spanish | 1357 | 1406 |
Instruction Following Kimi K2 Thinking Turbo leads
Claude Sonnet 4: 71.7 (#145), Kimi K2 Thinking Turbo: 74.0 (#109)
| Benchmark | Claude Sonnet 4 | Kimi K2 Thinking Turbo |
|---|---|---|
| LMArena Instruction Following | 1376 | 1403 |
| IFEval | 84% | — |
Long Context Kimi K2 Thinking Turbo leads
Claude Sonnet 4: 33.7 (#259), Kimi K2 Thinking Turbo: 43.2 (#102)
| Benchmark | Claude Sonnet 4 | Kimi K2 Thinking Turbo |
|---|---|---|
| LMArena Longer Query | 1398 | 1415 |
| Fiction.LiveBench | 46.9% | — |
Writing & Preference Kimi K2 Thinking Turbo leads
Claude Sonnet 4: 57.1 (#132), Kimi K2 Thinking Turbo: 60.0 (#104)
| Benchmark | Claude Sonnet 4 | Kimi K2 Thinking Turbo |
|---|---|---|
| LMArena Text | 1351 | 1415 |
| LMArena Creative Writing | 1345 | 1374 |
| LMArena Multi-Turn | 1376 | 1414 |
| Short-Story Creative Writing | 81.4% | — |
| EQ-Bench Creative Writing | 1483 | — |
| WildBench | 83.8% | — |
Frequently asked questions
Is Claude Sonnet 4 better than Kimi K2 Thinking Turbo?
Kimi K2 Thinking Turbo is the stronger model overall, scoring 45.8 to 40.8 on the Noometry Index.
Is Claude Sonnet 4 or Kimi K2 Thinking Turbo better for coding?
Claude Sonnet 4 scores higher on coding benchmarks: 43.5 versus 38.4 in the Noometry coding category.
How many benchmarks do Claude Sonnet 4 and Kimi K2 Thinking Turbo share?
19 benchmarks have published results for both models. Claude Sonnet 4 has 58 scored results on Noometry and Kimi K2 Thinking Turbo has 21.