Model comparison
Claude 3.7 Sonnet vs Kimi K3
Kimi K3 is the stronger model overall, scoring 59.5 to 39.5 on the Noometry Index.
Last verified . 25 shared benchmarks.
Summary
- They share 25 benchmarks with published results for both. Claude 3.7 Sonnet scores higher in 1 category and Kimi K3 in 9 categories; 10 gaps are clear of the uncertainty.
- The widest gap is in reasoning, where Kimi K3 leads 63.0 to 18.6.
- The biggest single-benchmark swing is ARC-AGI-1: 28.6% for Claude 3.7 Sonnet and 94.5% for Kimi K3.
- Kimi K3 has downloadable open weights; the other is API-only.
Side by side
| Claude 3.7 Sonnet | Kimi K3 | |
|---|---|---|
| Provider | Anthropic | Moonshot AI |
| Noometry Index | 39.5 | 59.5 |
| Released | 2025-02-24 | 2026-07-16 |
| Weights | Proprietary | Open |
| Context window | — | 1.05M |
| Max output | — | 1.05M |
| Input $ / M tokens | — | $3 |
| Output $ / M tokens | — | $15 |
| Results tracked | 58 | 53 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding Kimi K3 leads
Claude 3.7 Sonnet: 40.6 (#136), Kimi K3: 61.0 (#10)
| Benchmark | Claude 3.7 Sonnet | Kimi K3 |
|---|---|---|
| LMArena Coding | 1361 | 1508 |
| SWE-bench Verified | 61% | — |
| DeepSWE | — | 68.5% |
| FrontierCode | — | 44.2% |
| SWE-bench Verified (bash only) | 52.8% | — |
| Aider Polyglot | 64.9% | — |
| LMArena WebDev | — | 1654 |
| FrontierSWE | — | 25.9% |
| SciCode | — | 59.5% |
| GSO | 3.8% | — |
| WeirdML | — | 82.6% |
| LiveBench Coding | 74.5% | — |
| CadEval | 54% | — |
| ALE-Bench | — | 1,524 |
Agentic & Tool Use Kimi K3 leads
Claude 3.7 Sonnet: 34.1 (#50), Kimi K3: 41.8 (#20)
| Benchmark | Claude 3.7 Sonnet | Kimi K3 |
|---|---|---|
| APEX-Agents | — | 50.6% |
| TheAgentCompany | 30.9% | — |
| τ²-bench Banking | — | 37.1% |
| Cybench | 20% | — |
| DeepResearch Bench | 43.6% | — |
| OSWorld | 35.8% | — |
| PostTrainBench | — | 32% |
| GBAEval | — | 48.3% |
| GDP.pdf | — | 19% |
| METR Time Horizons | 60% | — |
| Vending-Bench 2 | — | 5,165 |
Reasoning Kimi K3 leads
Claude 3.7 Sonnet: 18.6 (#277), Kimi K3: 63.0 (#17)
| Benchmark | Claude 3.7 Sonnet | Kimi K3 |
|---|---|---|
| ARC-AGI-2 | 0.9% | 60.4% |
| SimpleBench | 46.4% | 60.7% |
| ARC-AGI-1 | 28.6% | 94.5% |
| LMArena Hard Prompts | 1333 | 1496 |
| Epoch Capabilities Index | 141.16 | 157.45 |
| ForecastBench | 61.8 | 61.1 |
| NYT Connections (extended) | — | 93.6% |
| CritPt | — | 23.4% |
| Chess Puzzles | — | 39% |
| EnigmaEval | 4.2% | — |
| LiveBench Reasoning | 87.8% | — |
| Mystery Game Puzzles | — | 26% |
| DTBench | — | 91.2% |
| LiveBench Data Analysis | 74% | — |
| LMCA | — | 52.7% |
| Surface Evolver Bench | — | 95% |
| LiveBench | 76.1% | — |
Math Kimi K3 leads
Claude 3.7 Sonnet: 37.5 (#153), Kimi K3: 74.2 (#16)
| Benchmark | Claude 3.7 Sonnet | Kimi K3 |
|---|---|---|
| OTIS Mock AIME 2024-2025 | 57.8% | 97.2% |
| LMArena Math | 1337 | 1491 |
| FrontierMath (Tiers 1-3) | — | 72.2% |
| FrontierMath Tier 4 | — | 39% |
| MathArena Final-Answer Competitions | — | 87.8% |
| ProofBench | — | 87% |
| Omni-MATH | 33% | — |
| LiveBench Math | 79% | — |
| MATH Level 5 | 91.2% | — |
| FrontierMath (Feb 2025 set) | 4.1% | — |
Knowledge Kimi K3 leads
Claude 3.7 Sonnet: 39.8 (#130), Kimi K3: 63.2 (#21)
| Benchmark | Claude 3.7 Sonnet | Kimi K3 |
|---|---|---|
| GPQA Diamond | 79.7% | 93.1% |
| LMArena Expert | 1321 | 1521 |
| Humanity's Last Exam | 8% | — |
| SimpleQA Verified | — | 50.6% |
| MMLU-Pro | 78.4% | — |
| Confabulations | 14.7% | — |
| GPQA (HELM) | 60.8% | — |
Multimodal Kimi K3 leads
Claude 3.7 Sonnet: 33.7 (#95), Kimi K3: 37.8 (#70)
| Benchmark | Claude 3.7 Sonnet | Kimi K3 |
|---|---|---|
| LMArena Vision | 1169 | — |
| GeoBench | 68% | — |
| VPCT | 39% | — |
| Blueprint-Bench 2 | — | 29.5% |
| Furniture Assembly | — | 34.2% |
| SpatialViz-Bench | 33.9% | — |
Multilingual Kimi K3 leads
Claude 3.7 Sonnet: 44.1 (#179), Kimi K3: 56.3 (#21)
| Benchmark | Claude 3.7 Sonnet | Kimi K3 |
|---|---|---|
| LMArena Non-English | 1296 | 1466 |
| LMArena Chinese | 1299 | 1529 |
| LMArena French | 1303 | 1491 |
| LMArena German | 1301 | 1488 |
| LMArena Japanese | 1267 | 1487 |
| LMArena Korean | 1249 | 1458 |
| LMArena Russian | 1311 | 1482 |
| LMArena Spanish | 1298 | 1472 |
Instruction Following Kimi K3 leads
Claude 3.7 Sonnet: 72.9 (#125), Kimi K3: 77.7 (#14)
| Benchmark | Claude 3.7 Sonnet | Kimi K3 |
|---|---|---|
| LMArena Instruction Following | 1352 | 1483 |
| LiveBench Instruction Following | 81.3% | — |
| IFEval | 83.4% | — |
Long Context Claude 3.7 Sonnet leads
Claude 3.7 Sonnet: 50.3 (#10), Kimi K3: 45.8 (#29)
| Benchmark | Claude 3.7 Sonnet | Kimi K3 |
|---|---|---|
| LMArena Longer Query | 1373 | 1494 |
| Fiction.LiveBench | 83.3% | — |
Writing & Preference Kimi K3 leads
Claude 3.7 Sonnet: 54.4 (#150), Kimi K3: 76.6 (#4)
| Benchmark | Claude 3.7 Sonnet | Kimi K3 |
|---|---|---|
| LMArena Text | 1314 | 1476 |
| LMArena Creative Writing | 1332 | 1454 |
| EQ-Bench Creative Writing | 1412 | 2082 |
| LMArena Multi-Turn | 1339 | 1488 |
| Short-Story Creative Writing | 81.1% | — |
| WildBench | 81.4% | — |
| EQ-Bench 4 | — | 1339 |
| LiveBench Language | 59.9% | — |
Frequently asked questions
Is Claude 3.7 Sonnet better than Kimi K3?
Kimi K3 is the stronger model overall, scoring 59.5 to 39.5 on the Noometry Index.
Is Claude 3.7 Sonnet or Kimi K3 better for coding?
Kimi K3 scores higher on coding benchmarks: 61.0 versus 40.6 in the Noometry coding category.
How many benchmarks do Claude 3.7 Sonnet and Kimi K3 share?
25 benchmarks have published results for both models. Claude 3.7 Sonnet has 58 scored results on Noometry and Kimi K3 has 53.