Model comparison
Claude Sonnet 4 vs Kimi K2 (Jul 2025)
Claude Sonnet 4 and Kimi K2 (Jul 2025) 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 . 39 shared benchmarks.
Summary
- They share 39 benchmarks with published results for both. Claude Sonnet 4 scores higher in 5 categories and Kimi K2 (Jul 2025) in 4 categories; 6 gaps are clear of the uncertainty.
- The widest gap is in long context, where Kimi K2 (Jul 2025) leads 41.2 to 33.7.
- The biggest single-benchmark swing is Fiction.LiveBench: 46.9% for Claude Sonnet 4 and 66.7% for Kimi K2 (Jul 2025).
- Kimi K2 (Jul 2025) is cheaper at $0.57 / $2.30 per million input/output tokens, against $3 / $15 for Claude Sonnet 4.
- Kimi K2 (Jul 2025) accepts more context: 262K tokens versus 200K.
- Kimi K2 (Jul 2025) has downloadable open weights; the other is API-only.
Side by side
| Claude Sonnet 4 | Kimi K2 (Jul 2025) | |
|---|---|---|
| Provider | Anthropic | Moonshot AI |
| Noometry Index | 40.8 | 41.2 |
| Released | 2025-05-22 | 2025-07-12 |
| Weights | Proprietary | Open |
| Context window | 200K | 262K |
| Max output | 64K | 262K |
| Input $ / M tokens | $3 | $0.57 |
| Output $ / M tokens | $15 | $2.30 |
| Results tracked | 58 | 42 |
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 (Jul 2025): 42.4 (#102)
| Benchmark | Claude Sonnet 4 | Kimi K2 (Jul 2025) |
|---|---|---|
| SWE-bench Verified (bash only) | 64.9% | 63.4% |
| Aider Polyglot | 61.3% | 59.1% |
| GSO | 4.9% | 4.9% |
| WeirdML | 46.1% | 42.8% |
| LMArena Coding | 1414 | 1399 |
| ALE-Bench | 655.35 | 597.5 |
| SciCode | 40% | — |
Agentic & Tool Use Claude Sonnet 4 leads
Claude Sonnet 4: 38.5 (#31), Kimi K2 (Jul 2025): 32.4 (#64)
| Benchmark | Claude Sonnet 4 | Kimi K2 (Jul 2025) |
|---|---|---|
| METR Time Horizons | 62% | 59.2% |
| Terminal-Bench | — | 35.7% |
| Berkeley Function Calling Leaderboard | — | 59.1% |
| TheAgentCompany | 33.1% | — |
| Cybench | 35% | — |
| DeepResearch Bench | 46.6% | — |
| OSWorld | 43.9% | — |
Reasoning Too close to call
Claude Sonnet 4: 22.9 (#187), Kimi K2 (Jul 2025): 23.3 (#179)
| Benchmark | Claude Sonnet 4 | Kimi K2 (Jul 2025) |
|---|---|---|
| SimpleBench | 45.5% | 26.3% |
| Kagi LLM Benchmark | 73% | 64.4% |
| LMArena Hard Prompts | 1372 | 1384 |
| Epoch Capabilities Index | 141.69 | 146.01 |
| ForecastBench | 60.2 | 60.2 |
| ARC-AGI-2 | 5.9% | — |
| ARC-AGI-1 | 40% | — |
| CritPt | 0.3% | — |
| EnigmaEval | 3.1% | — |
| DTBench | 77.1% | — |
| LMCA | 29% | — |
Math Too close to call
Claude Sonnet 4: 43.3 (#80), Kimi K2 (Jul 2025): 42.7 (#83)
| Benchmark | Claude Sonnet 4 | Kimi K2 (Jul 2025) |
|---|---|---|
| Omni-MATH | 60.2% | 65.4% |
| LMArena Math | 1375 | 1397 |
| FrontierMath (Feb 2025 set) | 4.1% | 21.4% |
| FrontierMath Tier 4 (v1) | 0% | 0% |
| OTIS Mock AIME 2024-2025 | 71.1% | — |
| MATH Level 5 | 84.4% | — |
Knowledge Claude Sonnet 4 leads
Claude Sonnet 4: 41.8 (#108), Kimi K2 (Jul 2025): 37.3 (#157)
| Benchmark | Claude Sonnet 4 | Kimi K2 (Jul 2025) |
|---|---|---|
| MMLU-Pro | 84.3% | 81.9% |
| Confabulations | 13.2% | 20.4% |
| Vectara Hallucination Rate | 10.3% | 17.9% |
| GPQA (HELM) | 70.6% | 65.3% |
| LMArena Expert | 1372 | 1365 |
| GPQA Diamond | 79.2% | — |
| Humanity's Last Exam | 7.8% | — |
Multimodal Not comparable
Claude Sonnet 4: 26.2 (#121), Kimi K2 (Jul 2025): —
| Benchmark | Claude Sonnet 4 | Kimi K2 (Jul 2025) |
|---|---|---|
| LMArena Vision | 1191 | — |
| GeoBench | 37% | — |
| VPCT | 34% | — |
| MindCube | 44.8% | — |
Multilingual Kimi K2 (Jul 2025) leads
Claude Sonnet 4: 46.7 (#156), Kimi K2 (Jul 2025): 49.6 (#130)
| Benchmark | Claude Sonnet 4 | Kimi K2 (Jul 2025) |
|---|---|---|
| LMArena Non-English | 1333 | 1372 |
| LMArena Chinese | 1350 | 1415 |
| LMArena French | 1363 | 1379 |
| LMArena German | 1331 | 1387 |
| LMArena Japanese | 1302 | 1349 |
| LMArena Korean | 1291 | 1325 |
| LMArena Russian | 1355 | 1385 |
| LMArena Spanish | 1357 | 1386 |
Instruction Following Too close to call
Claude Sonnet 4: 71.7 (#145), Kimi K2 (Jul 2025): 71.1 (#156)
| Benchmark | Claude Sonnet 4 | Kimi K2 (Jul 2025) |
|---|---|---|
| IFEval | 84% | 85% |
| LMArena Instruction Following | 1376 | 1348 |
Long Context Kimi K2 (Jul 2025) leads
Claude Sonnet 4: 33.7 (#259), Kimi K2 (Jul 2025): 41.2 (#145)
| Benchmark | Claude Sonnet 4 | Kimi K2 (Jul 2025) |
|---|---|---|
| Fiction.LiveBench | 46.9% | 66.7% |
| LMArena Longer Query | 1398 | 1353 |
| CL-bench | — | 17.6% |
Writing & Preference Kimi K2 (Jul 2025) leads
Claude Sonnet 4: 57.1 (#132), Kimi K2 (Jul 2025): 62.3 (#78)
| Benchmark | Claude Sonnet 4 | Kimi K2 (Jul 2025) |
|---|---|---|
| LMArena Text | 1351 | 1380 |
| LMArena Creative Writing | 1345 | 1350 |
| Short-Story Creative Writing | 81.4% | 85.6% |
| EQ-Bench Creative Writing | 1483 | 1666 |
| WildBench | 83.8% | 86.2% |
| LMArena Multi-Turn | 1376 | 1371 |
Frequently asked questions
Is Claude Sonnet 4 better than Kimi K2 (Jul 2025)?
Claude Sonnet 4 and Kimi K2 (Jul 2025) 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.
Which is cheaper, Claude Sonnet 4 or Kimi K2 (Jul 2025)?
Kimi K2 (Jul 2025) is cheaper. It lists at $0.57 per million input tokens and $2.30 per million output tokens; Claude Sonnet 4 lists at $3 and $15.
Is Claude Sonnet 4 or Kimi K2 (Jul 2025) better for coding?
Claude Sonnet 4 scores higher on coding benchmarks: 43.5 versus 42.4 in the Noometry coding category.
Which has the bigger context window?
Kimi K2 (Jul 2025) does, with 262K tokens against 200K.
How many benchmarks do Claude Sonnet 4 and Kimi K2 (Jul 2025) share?
39 benchmarks have published results for both models. Claude Sonnet 4 has 58 scored results on Noometry and Kimi K2 (Jul 2025) has 42.