Model comparison
Claude 3.7 Sonnet vs Kimi K2.5
Kimi K2.5 is the stronger model overall, scoring 48.1 to 39.5 on the Noometry Index.
Last verified . 32 shared benchmarks.
Summary
- They share 32 benchmarks with published results for both. Claude 3.7 Sonnet scores higher in 0 categories and Kimi K2.5 in 10 categories; 9 gaps are clear of the uncertainty.
- The widest gap is in math, where Kimi K2.5 leads 51.8 to 37.5.
- The biggest single-benchmark swing is ARC-AGI-1: 28.6% for Claude 3.7 Sonnet and 65.3% for Kimi K2.5.
- Kimi K2.5 has downloadable open weights; the other is API-only.
Side by side
| Claude 3.7 Sonnet | Kimi K2.5 | |
|---|---|---|
| Provider | Anthropic | Moonshot AI |
| Noometry Index | 39.5 | 48.1 |
| Released | 2025-02-24 | 2026-01-27 |
| Weights | Proprietary | Open |
| Context window | — | 262K |
| Max output | — | 262K |
| Input $ / M tokens | — | $0.45 |
| Output $ / M tokens | — | $2.25 |
| Results tracked | 58 | 51 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding Kimi K2.5 leads
Claude 3.7 Sonnet: 40.6 (#136), Kimi K2.5: 48.8 (#53)
| Benchmark | Claude 3.7 Sonnet | Kimi K2.5 |
|---|---|---|
| SWE-bench Verified | 61% | 73.8% |
| SWE-bench Verified (bash only) | 52.8% | 70.8% |
| LMArena Coding | 1361 | 1474 |
| Aider Polyglot | 64.9% | — |
| LMArena WebDev | — | 1437 |
| SWE-bench Multilingual | — | 67.3% |
| SciCode | — | 49% |
| GSO | 3.8% | — |
| WeirdML | — | 45.6% |
| LiveBench Coding | 74.5% | — |
| CadEval | 54% | — |
| ALE-Bench | — | 821.65 |
Agentic & Tool Use Too close to call
Claude 3.7 Sonnet: 34.1 (#50), Kimi K2.5: 34.2 (#48)
| Benchmark | Claude 3.7 Sonnet | Kimi K2.5 |
|---|---|---|
| OSWorld | 35.8% | 63.3% |
| Terminal-Bench | — | 43.2% |
| TheAgentCompany | 30.9% | — |
| Cybench | 20% | — |
| DeepResearch Bench | 43.6% | — |
| METR Time Horizons | 60% | — |
| Vending-Bench 2 | — | 1,198 |
Reasoning Kimi K2.5 leads
Claude 3.7 Sonnet: 18.6 (#277), Kimi K2.5: 31.2 (#80)
| Benchmark | Claude 3.7 Sonnet | Kimi K2.5 |
|---|---|---|
| ARC-AGI-2 | 0.9% | 11.8% |
| SimpleBench | 46.4% | 46.8% |
| ARC-AGI-1 | 28.6% | 65.3% |
| EnigmaEval | 4.2% | 3.4% |
| LMArena Hard Prompts | 1333 | 1453 |
| Epoch Capabilities Index | 141.16 | 148.03 |
| Kagi LLM Benchmark | — | 78.5% |
| NYT Connections (extended) | — | 69.9% |
| CritPt | — | 3.1% |
| Chess Puzzles | — | 12% |
| Thematic Generalization | — | 69.4% |
| LiveBench Reasoning | 87.8% | — |
| LiveBench Data Analysis | 74% | — |
| ForecastBench | 61.8 | — |
| LiveBench | 76.1% | — |
Math Kimi K2.5 leads
Claude 3.7 Sonnet: 37.5 (#153), Kimi K2.5: 51.8 (#53)
| Benchmark | Claude 3.7 Sonnet | Kimi K2.5 |
|---|---|---|
| OTIS Mock AIME 2024-2025 | 57.8% | 92.2% |
| LMArena Math | 1337 | 1470 |
| FrontierMath (Feb 2025 set) | 4.1% | 27.9% |
| MathArena Final-Answer Competitions | — | 62.3% |
| Omni-MATH | 33% | — |
| LiveBench Math | 79% | — |
| MATH Level 5 | 91.2% | — |
| FrontierMath Tier 4 (v1) | — | 4.2% |
Knowledge Kimi K2.5 leads
Claude 3.7 Sonnet: 39.8 (#130), Kimi K2.5: 53.6 (#56)
| Benchmark | Claude 3.7 Sonnet | Kimi K2.5 |
|---|---|---|
| GPQA Diamond | 79.7% | 87.6% |
| Humanity's Last Exam | 8% | 24.4% |
| LMArena Expert | 1321 | 1466 |
| SimpleQA Verified | — | 34.3% |
| MMLU-Pro | 78.4% | — |
| Confabulations | 14.7% | — |
| Vectara Hallucination Rate | — | 14.2% |
| GPQA (HELM) | 60.8% | — |
Multimodal Kimi K2.5 leads
Claude 3.7 Sonnet: 33.7 (#95), Kimi K2.5: 41.1 (#39)
| Benchmark | Claude 3.7 Sonnet | Kimi K2.5 |
|---|---|---|
| LMArena Vision | 1169 | 1269 |
| GeoBench | 68% | — |
| VPCT | 39% | — |
| LMArena Document | — | 1430 |
| SpatialViz-Bench | 33.9% | — |
Multilingual Kimi K2.5 leads
Claude 3.7 Sonnet: 44.1 (#179), Kimi K2.5: 53.9 (#53)
| Benchmark | Claude 3.7 Sonnet | Kimi K2.5 |
|---|---|---|
| LMArena Non-English | 1296 | 1433 |
| LMArena Chinese | 1299 | 1495 |
| LMArena French | 1303 | 1454 |
| LMArena German | 1301 | 1441 |
| LMArena Japanese | 1267 | 1421 |
| LMArena Korean | 1249 | 1410 |
| LMArena Russian | 1311 | 1435 |
| LMArena Spanish | 1298 | 1450 |
Instruction Following Kimi K2.5 leads
Claude 3.7 Sonnet: 72.9 (#125), Kimi K2.5: 75.3 (#64)
| Benchmark | Claude 3.7 Sonnet | Kimi K2.5 |
|---|---|---|
| LMArena Instruction Following | 1352 | 1431 |
| LiveBench Instruction Following | 81.3% | — |
| IFEval | 83.4% | — |
Long Context Kimi K2.5 leads
Claude 3.7 Sonnet: 50.3 (#10), Kimi K2.5: 52.1 (#7)
| Benchmark | Claude 3.7 Sonnet | Kimi K2.5 |
|---|---|---|
| Fiction.LiveBench | 83.3% | 86.1% |
| LMArena Longer Query | 1373 | 1445 |
| CL-bench | — | 19.3% |
| CL-bench Life | — | 13.2% |
Writing & Preference Kimi K2.5 leads
Claude 3.7 Sonnet: 54.4 (#150), Kimi K2.5: 65.1 (#53)
| Benchmark | Claude 3.7 Sonnet | Kimi K2.5 |
|---|---|---|
| LMArena Text | 1314 | 1445 |
| LMArena Creative Writing | 1332 | 1423 |
| EQ-Bench Creative Writing | 1412 | 1579 |
| LMArena Multi-Turn | 1339 | 1444 |
| Short-Story Creative Writing | 81.1% | — |
| WildBench | 81.4% | — |
| LiveBench Language | 59.9% | — |
Frequently asked questions
Is Claude 3.7 Sonnet better than Kimi K2.5?
Kimi K2.5 is the stronger model overall, scoring 48.1 to 39.5 on the Noometry Index.
Is Claude 3.7 Sonnet or Kimi K2.5 better for coding?
Kimi K2.5 scores higher on coding benchmarks: 48.8 versus 40.6 in the Noometry coding category.
How many benchmarks do Claude 3.7 Sonnet and Kimi K2.5 share?
32 benchmarks have published results for both models. Claude 3.7 Sonnet has 58 scored results on Noometry and Kimi K2.5 has 51.