Model comparison
Claude Sonnet 4.5 vs Kimi K3
Kimi K3 is the stronger model overall, scoring 59.5 to 44.1 on the Noometry Index.
Last verified . 41 shared benchmarks.
Summary
- They share 41 benchmarks with published results for both. Claude Sonnet 4.5 scores higher in 0 categories and Kimi K3 in 10 categories; 9 gaps are clear of the uncertainty.
- The widest gap is in math, where Kimi K3 leads 74.2 to 32.3.
- The biggest single-benchmark swing is ProofBench: 19% for Claude Sonnet 4.5 and 87% for Kimi K3.
- Both cost about the same: $3 input and $15 output per million tokens.
- Kimi K3 accepts more context: 1.05M tokens versus 200K.
- Kimi K3 has downloadable open weights; the other is API-only.
Side by side
| Claude Sonnet 4.5 | Kimi K3 | |
|---|---|---|
| Provider | Anthropic | Moonshot AI |
| Noometry Index | 44.1 | 59.5 |
| Released | 2025-09-29 | 2026-07-16 |
| Weights | Proprietary | Open |
| Context window | 200K | 1.05M |
| Max output | 64K | 1.05M |
| Input $ / M tokens | $3 | $3 |
| Output $ / M tokens | $15 | $15 |
| Results tracked | 73 | 53 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding Kimi K3 leads
Claude Sonnet 4.5: 47.3 (#61), Kimi K3: 61.0 (#10)
| Benchmark | Claude Sonnet 4.5 | Kimi K3 |
|---|---|---|
| LMArena WebDev | 1393 | 1654 |
| SciCode | 44.7% | 59.5% |
| WeirdML | 47.7% | 82.6% |
| LMArena Coding | 1489 | 1508 |
| ALE-Bench | 796.15 | 1,524 |
| SWE-bench Verified | 71.3% | — |
| DeepSWE | — | 68.5% |
| FrontierCode | — | 44.2% |
| SWE-bench Verified (bash only) | 71.4% | — |
| SWE-bench Multilingual | 67% | — |
| FrontierSWE | — | 25.9% |
| GSO | 14.7% | — |
| AlgoTune | 1.52 | — |
Agentic & Tool Use Kimi K3 leads
Claude Sonnet 4.5: 38.3 (#32), Kimi K3: 41.8 (#20)
| Benchmark | Claude Sonnet 4.5 | Kimi K3 |
|---|---|---|
| τ²-bench Banking | 25.3% | 37.1% |
| Vending-Bench 2 | 3,839 | 5,165 |
| Terminal-Bench | 46.5% | — |
| APEX-Agents | — | 50.6% |
| Berkeley Function Calling Leaderboard | 73.2% | — |
| GDPval | 42.5% | — |
| Remote Labor Index | 2.1% | — |
| τ²-bench Airline | 72% | — |
| τ²-bench Retail | 72.4% | — |
| τ²-bench Telecom | 84.9% | — |
| Cybench | 60% | — |
| DeepResearch Bench | 52.6% | — |
| OSWorld | 62.9% | — |
| PostTrainBench | — | 32% |
| GBAEval | — | 48.3% |
| GDP.pdf | — | 19% |
| LMArena Search | 1159 | — |
| METR Time Horizons | 67.4% | — |
Reasoning Kimi K3 leads
Claude Sonnet 4.5: 26.9 (#125), Kimi K3: 63.0 (#17)
| Benchmark | Claude Sonnet 4.5 | Kimi K3 |
|---|---|---|
| ARC-AGI-2 | 13.6% | 60.4% |
| SimpleBench | 54.3% | 60.7% |
| NYT Connections (extended) | 37.3% | 93.6% |
| ARC-AGI-1 | 63.7% | 94.5% |
| CritPt | 1.1% | 23.4% |
| Chess Puzzles | 12% | 39% |
| LMArena Hard Prompts | 1462 | 1496 |
| Mystery Game Puzzles | 17% | 26% |
| DTBench | 83.2% | 91.2% |
| LMCA | 38.8% | 52.7% |
| Epoch Capabilities Index | 146.84 | 157.45 |
| ForecastBench | 61.9 | 61.1 |
| Kagi LLM Benchmark | 57.9% | — |
| EnigmaEval | 6% | — |
| EBR-Bench | 2.4% | — |
| Surface Evolver Bench | — | 95% |
Math Kimi K3 leads
Claude Sonnet 4.5: 32.3 (#216), Kimi K3: 74.2 (#16)
| Benchmark | Claude Sonnet 4.5 | Kimi K3 |
|---|---|---|
| FrontierMath (Tiers 1-3) | 23.9% | 72.2% |
| FrontierMath Tier 4 | 2.4% | 39% |
| OTIS Mock AIME 2024-2025 | 77.8% | 97.2% |
| ProofBench | 19% | 87% |
| LMArena Math | 1449 | 1491 |
| MathArena Final-Answer Competitions | — | 87.8% |
| Omni-MATH | 55.3% | — |
| MATH Level 5 | 97.7% | — |
| FrontierMath (Feb 2025 set) | 15.2% | — |
| FrontierMath Tier 4 (v1) | 4.2% | — |
Knowledge Kimi K3 leads
Claude Sonnet 4.5: 48.4 (#76), Kimi K3: 63.2 (#21)
| Benchmark | Claude Sonnet 4.5 | Kimi K3 |
|---|---|---|
| GPQA Diamond | 82.3% | 93.1% |
| SimpleQA Verified | 30.7% | 50.6% |
| LMArena Expert | 1482 | 1521 |
| Humanity's Last Exam | 13.7% | — |
| MMLU-Pro | 86.9% | — |
| Vectara Hallucination Rate | 12% | — |
| GPQA (HELM) | 68.6% | — |
Multimodal Kimi K3 leads
Claude Sonnet 4.5: 34.8 (#89), Kimi K3: 37.8 (#70)
| Benchmark | Claude Sonnet 4.5 | Kimi K3 |
|---|---|---|
| VPCT | 39.8% | — |
| Blueprint-Bench 2 | — | 29.5% |
| Furniture Assembly | — | 34.2% |
| LMArena Document | 1450 | — |
Multilingual Kimi K3 leads
Claude Sonnet 4.5: 53.4 (#69), Kimi K3: 56.3 (#21)
| Benchmark | Claude Sonnet 4.5 | Kimi K3 |
|---|---|---|
| LMArena Non-English | 1425 | 1466 |
| LMArena Chinese | 1459 | 1529 |
| LMArena French | 1458 | 1491 |
| LMArena German | 1427 | 1488 |
| LMArena Japanese | 1390 | 1487 |
| LMArena Korean | 1403 | 1458 |
| LMArena Russian | 1437 | 1482 |
| LMArena Spanish | 1457 | 1472 |
Instruction Following Kimi K3 leads
Claude Sonnet 4.5: 75.0 (#78), Kimi K3: 77.7 (#14)
| Benchmark | Claude Sonnet 4.5 | Kimi K3 |
|---|---|---|
| LMArena Instruction Following | 1459 | 1483 |
| IFEval | 85% | — |
Long Context Too close to call
Claude Sonnet 4.5: 45.2 (#46), Kimi K3: 45.8 (#29)
| Benchmark | Claude Sonnet 4.5 | Kimi K3 |
|---|---|---|
| LMArena Longer Query | 1476 | 1494 |
Writing & Preference Kimi K3 leads
Claude Sonnet 4.5: 66.5 (#34), Kimi K3: 76.6 (#4)
| Benchmark | Claude Sonnet 4.5 | Kimi K3 |
|---|---|---|
| LMArena Text | 1439 | 1476 |
| LMArena Creative Writing | 1442 | 1454 |
| EQ-Bench Creative Writing | 1678 | 2082 |
| LMArena Multi-Turn | 1465 | 1488 |
| WildBench | 85.4% | — |
| EQ-Bench 4 | — | 1339 |
Frequently asked questions
Is Claude Sonnet 4.5 better than Kimi K3?
Kimi K3 is the stronger model overall, scoring 59.5 to 44.1 on the Noometry Index.
Which is cheaper, Claude Sonnet 4.5 or Kimi K3?
Kimi K3 is cheaper. It lists at $3 per million input tokens and $15 per million output tokens; Claude Sonnet 4.5 lists at $3 and $15.
Is Claude Sonnet 4.5 or Kimi K3 better for coding?
Kimi K3 scores higher on coding benchmarks: 61.0 versus 47.3 in the Noometry coding category.
Which has the bigger context window?
Kimi K3 does, with 1.05M tokens against 200K.
How many benchmarks do Claude Sonnet 4.5 and Kimi K3 share?
41 benchmarks have published results for both models. Claude Sonnet 4.5 has 73 scored results on Noometry and Kimi K3 has 53.