Model comparison
Claude Sonnet 4.5 vs Kimi K2.5
Kimi K2.5 is the stronger model overall, scoring 48.1 to 44.1 on the Noometry Index.
Last verified . 45 shared benchmarks.
Summary
- They share 45 benchmarks with published results for both. Claude Sonnet 4.5 scores higher in 2 categories and Kimi K2.5 in 8 categories; 8 gaps are clear of the uncertainty.
- The widest gap is in math, where Kimi K2.5 leads 51.8 to 32.3.
- The biggest single-benchmark swing is NYT Connections (extended): 37.3% for Claude Sonnet 4.5 and 69.9% for Kimi K2.5.
- Kimi K2.5 is cheaper at $0.45 / $2.25 per million input/output tokens, against $3 / $15 for Claude Sonnet 4.5.
- Kimi K2.5 accepts more context: 262K tokens versus 200K.
- Kimi K2.5 has downloadable open weights; the other is API-only.
Side by side
| Claude Sonnet 4.5 | Kimi K2.5 | |
|---|---|---|
| Provider | Anthropic | Moonshot AI |
| Noometry Index | 44.1 | 48.1 |
| Released | 2025-09-29 | 2026-01-27 |
| Weights | Proprietary | Open |
| Context window | 200K | 262K |
| Max output | 64K | 262K |
| Input $ / M tokens | $3 | $0.45 |
| Output $ / M tokens | $15 | $2.25 |
| Results tracked | 73 | 51 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding Kimi K2.5 leads
Claude Sonnet 4.5: 47.3 (#61), Kimi K2.5: 48.8 (#53)
| Benchmark | Claude Sonnet 4.5 | Kimi K2.5 |
|---|---|---|
| SWE-bench Verified | 71.3% | 73.8% |
| SWE-bench Verified (bash only) | 71.4% | 70.8% |
| LMArena WebDev | 1393 | 1437 |
| SWE-bench Multilingual | 67% | 67.3% |
| SciCode | 44.7% | 49% |
| WeirdML | 47.7% | 45.6% |
| LMArena Coding | 1489 | 1474 |
| ALE-Bench | 796.15 | 821.65 |
| GSO | 14.7% | — |
| AlgoTune | 1.52 | — |
Agentic & Tool Use Claude Sonnet 4.5 leads
Claude Sonnet 4.5: 38.3 (#32), Kimi K2.5: 34.2 (#48)
| Benchmark | Claude Sonnet 4.5 | Kimi K2.5 |
|---|---|---|
| Terminal-Bench | 46.5% | 43.2% |
| OSWorld | 62.9% | 63.3% |
| Vending-Bench 2 | 3,839 | 1,198 |
| Berkeley Function Calling Leaderboard | 73.2% | — |
| GDPval | 42.5% | — |
| Remote Labor Index | 2.1% | — |
| τ²-bench Airline | 72% | — |
| τ²-bench Banking | 25.3% | — |
| τ²-bench Retail | 72.4% | — |
| τ²-bench Telecom | 84.9% | — |
| Cybench | 60% | — |
| DeepResearch Bench | 52.6% | — |
| LMArena Search | 1159 | — |
| METR Time Horizons | 67.4% | — |
Reasoning Kimi K2.5 leads
Claude Sonnet 4.5: 26.9 (#125), Kimi K2.5: 31.2 (#80)
| Benchmark | Claude Sonnet 4.5 | Kimi K2.5 |
|---|---|---|
| ARC-AGI-2 | 13.6% | 11.8% |
| SimpleBench | 54.3% | 46.8% |
| Kagi LLM Benchmark | 57.9% | 78.5% |
| NYT Connections (extended) | 37.3% | 69.9% |
| ARC-AGI-1 | 63.7% | 65.3% |
| CritPt | 1.1% | 3.1% |
| Chess Puzzles | 12% | 12% |
| EnigmaEval | 6% | 3.4% |
| LMArena Hard Prompts | 1462 | 1453 |
| Epoch Capabilities Index | 146.84 | 148.03 |
| Thematic Generalization | — | 69.4% |
| EBR-Bench | 2.4% | — |
| Mystery Game Puzzles | 17% | — |
| DTBench | 83.2% | — |
| LMCA | 38.8% | — |
| ForecastBench | 61.9 | — |
Math Kimi K2.5 leads
Claude Sonnet 4.5: 32.3 (#216), Kimi K2.5: 51.8 (#53)
| Benchmark | Claude Sonnet 4.5 | Kimi K2.5 |
|---|---|---|
| OTIS Mock AIME 2024-2025 | 77.8% | 92.2% |
| LMArena Math | 1449 | 1470 |
| FrontierMath (Feb 2025 set) | 15.2% | 27.9% |
| FrontierMath Tier 4 (v1) | 4.2% | 4.2% |
| FrontierMath (Tiers 1-3) | 23.9% | — |
| FrontierMath Tier 4 | 2.4% | — |
| MathArena Final-Answer Competitions | — | 62.3% |
| ProofBench | 19% | — |
| Omni-MATH | 55.3% | — |
| MATH Level 5 | 97.7% | — |
Knowledge Kimi K2.5 leads
Claude Sonnet 4.5: 48.4 (#76), Kimi K2.5: 53.6 (#56)
| Benchmark | Claude Sonnet 4.5 | Kimi K2.5 |
|---|---|---|
| GPQA Diamond | 82.3% | 87.6% |
| Humanity's Last Exam | 13.7% | 24.4% |
| SimpleQA Verified | 30.7% | 34.3% |
| Vectara Hallucination Rate | 12% | 14.2% |
| LMArena Expert | 1482 | 1466 |
| MMLU-Pro | 86.9% | — |
| GPQA (HELM) | 68.6% | — |
Multimodal Kimi K2.5 leads
Claude Sonnet 4.5: 34.8 (#89), Kimi K2.5: 41.1 (#39)
| Benchmark | Claude Sonnet 4.5 | Kimi K2.5 |
|---|---|---|
| LMArena Document | 1450 | 1430 |
| LMArena Vision | — | 1269 |
| VPCT | 39.8% | — |
Multilingual Too close to call
Claude Sonnet 4.5: 53.4 (#69), Kimi K2.5: 53.9 (#53)
| Benchmark | Claude Sonnet 4.5 | Kimi K2.5 |
|---|---|---|
| LMArena Non-English | 1425 | 1433 |
| LMArena Chinese | 1459 | 1495 |
| LMArena French | 1458 | 1454 |
| LMArena German | 1427 | 1441 |
| LMArena Japanese | 1390 | 1421 |
| LMArena Korean | 1403 | 1410 |
| LMArena Russian | 1437 | 1435 |
| LMArena Spanish | 1457 | 1450 |
Instruction Following Too close to call
Claude Sonnet 4.5: 75.0 (#78), Kimi K2.5: 75.3 (#64)
| Benchmark | Claude Sonnet 4.5 | Kimi K2.5 |
|---|---|---|
| LMArena Instruction Following | 1459 | 1431 |
| IFEval | 85% | — |
Long Context Kimi K2.5 leads
Claude Sonnet 4.5: 45.2 (#46), Kimi K2.5: 52.1 (#7)
| Benchmark | Claude Sonnet 4.5 | Kimi K2.5 |
|---|---|---|
| LMArena Longer Query | 1476 | 1445 |
| Fiction.LiveBench | — | 86.1% |
| CL-bench | — | 19.3% |
| CL-bench Life | — | 13.2% |
Writing & Preference Claude Sonnet 4.5 leads
Claude Sonnet 4.5: 66.5 (#34), Kimi K2.5: 65.1 (#53)
| Benchmark | Claude Sonnet 4.5 | Kimi K2.5 |
|---|---|---|
| LMArena Text | 1439 | 1445 |
| LMArena Creative Writing | 1442 | 1423 |
| EQ-Bench Creative Writing | 1678 | 1579 |
| LMArena Multi-Turn | 1465 | 1444 |
| WildBench | 85.4% | — |
Frequently asked questions
Is Claude Sonnet 4.5 better than Kimi K2.5?
Kimi K2.5 is the stronger model overall, scoring 48.1 to 44.1 on the Noometry Index.
Which is cheaper, Claude Sonnet 4.5 or Kimi K2.5?
Kimi K2.5 is cheaper. It lists at $0.45 per million input tokens and $2.25 per million output tokens; Claude Sonnet 4.5 lists at $3 and $15.
Is Claude Sonnet 4.5 or Kimi K2.5 better for coding?
Kimi K2.5 scores higher on coding benchmarks: 48.8 versus 47.3 in the Noometry coding category.
Which has the bigger context window?
Kimi K2.5 does, with 262K tokens against 200K.
How many benchmarks do Claude Sonnet 4.5 and Kimi K2.5 share?
45 benchmarks have published results for both models. Claude Sonnet 4.5 has 73 scored results on Noometry and Kimi K2.5 has 51.