Model comparison
Claude Sonnet 4.6 vs Kimi K2.5
Claude Sonnet 4.6 is the stronger model overall, scoring 50.3 to 48.1 on the Noometry Index. Kimi K2.5 costs 6.7× less per token, which makes it the better buy when Claude Sonnet 4.6's lead doesn't matter for your workload.
Last verified . 41 shared benchmarks.
Summary
- They share 41 benchmarks with published results for both. Claude Sonnet 4.6 scores higher in 6 categories and Kimi K2.5 in 4 categories; 9 gaps are clear of the uncertainty.
- The widest gap is in reasoning, where Claude Sonnet 4.6 leads 46.1 to 31.2.
- The biggest single-benchmark swing is ARC-AGI-2: 60.4% for Claude Sonnet 4.6 and 11.8% 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 4.6.
- Claude Sonnet 4.6 accepts more context: 1M tokens versus 262K.
- Kimi K2.5 has downloadable open weights; the other is API-only.
Side by side
| Claude Sonnet 4.6 | Kimi K2.5 | |
|---|---|---|
| Provider | Anthropic | Moonshot AI |
| Noometry Index | 50.3 | 48.1 |
| Released | 2026-02-17 | 2026-01-27 |
| Weights | Proprietary | Open |
| Context window | 1M | 262K |
| Max output | 128K | 262K |
| Input $ / M tokens | $3 | $0.45 |
| Output $ / M tokens | $15 | $2.25 |
| Results tracked | 57 | 51 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding Kimi K2.5 leads
Claude Sonnet 4.6: 46.3 (#67), Kimi K2.5: 48.8 (#53)
| Benchmark | Claude Sonnet 4.6 | Kimi K2.5 |
|---|---|---|
| SWE-bench Verified | 75.2% | 73.8% |
| LMArena WebDev | 1522 | 1437 |
| SciCode | 46.8% | 49% |
| WeirdML | 66.1% | 45.6% |
| LMArena Coding | 1504 | 1474 |
| ALE-Bench | 1,327 | 821.65 |
| DeepSWE | 29.9% | — |
| FrontierCode | 24.3% | — |
| SWE-bench Verified (bash only) | — | 70.8% |
| SWE-bench Multilingual | — | 67.3% |
Agentic & Tool Use Claude Sonnet 4.6 leads
Claude Sonnet 4.6: 39.1 (#28), Kimi K2.5: 34.2 (#48)
| Benchmark | Claude Sonnet 4.6 | Kimi K2.5 |
|---|---|---|
| Terminal-Bench | 53.4% | 43.2% |
| OSWorld | 72.1% | 63.3% |
| Vending-Bench 2 | 7,204 | 1,198 |
| APEX-Agents | 43% | — |
| OSWorld 2.0 | 9.3% | — |
| DeepResearch Bench | 54.9% | — |
| ExploitBench | 23.6% | — |
| GBAEval | 48.8% | — |
| GDP.pdf | 18% | — |
| LMArena Search | 1221 | — |
Reasoning Claude Sonnet 4.6 leads
Claude Sonnet 4.6: 46.1 (#45), Kimi K2.5: 31.2 (#80)
| Benchmark | Claude Sonnet 4.6 | Kimi K2.5 |
|---|---|---|
| ARC-AGI-2 | 60.4% | 11.8% |
| NYT Connections (extended) | 80.9% | 69.9% |
| ARC-AGI-1 | 86.5% | 65.3% |
| CritPt | 3.1% | 3.1% |
| Chess Puzzles | 13% | 12% |
| Thematic Generalization | 76.3% | 69.4% |
| LMArena Hard Prompts | 1484 | 1453 |
| Epoch Capabilities Index | 152.24 | 148.03 |
| SimpleBench | — | 46.8% |
| Kagi LLM Benchmark | — | 78.5% |
| EnigmaEval | — | 3.4% |
| Mystery Game Puzzles | 16% | — |
| DTBench | 89.9% | — |
| LMCA | 46.5% | — |
| ForecastBench | 62 | — |
Math Claude Sonnet 4.6 leads
Claude Sonnet 4.6: 52.9 (#49), Kimi K2.5: 51.8 (#53)
| Benchmark | Claude Sonnet 4.6 | Kimi K2.5 |
|---|---|---|
| OTIS Mock AIME 2024-2025 | 85.8% | 92.2% |
| LMArena Math | 1462 | 1470 |
| FrontierMath (Feb 2025 set) | 32.4% | 27.9% |
| FrontierMath Tier 4 (v1) | 8.3% | 4.2% |
| MathArena Final-Answer Competitions | — | 62.3% |
| ProofBench | 45% | — |
Knowledge Kimi K2.5 leads
Claude Sonnet 4.6: 51.7 (#65), Kimi K2.5: 53.6 (#56)
| Benchmark | Claude Sonnet 4.6 | Kimi K2.5 |
|---|---|---|
| GPQA Diamond | 87.4% | 87.6% |
| SimpleQA Verified | 35.5% | 34.3% |
| Vectara Hallucination Rate | 10.6% | 14.2% |
| LMArena Expert | 1500 | 1466 |
| Humanity's Last Exam | — | 24.4% |
Multimodal Kimi K2.5 leads
Claude Sonnet 4.6: 38.0 (#68), Kimi K2.5: 41.1 (#39)
| Benchmark | Claude Sonnet 4.6 | Kimi K2.5 |
|---|---|---|
| LMArena Vision | 1283 | 1269 |
| LMArena Document | 1482 | 1430 |
| Blueprint-Bench 2 | 6.7% | — |
Multilingual Too close to call
Claude Sonnet 4.6: 54.4 (#41), Kimi K2.5: 53.9 (#53)
| Benchmark | Claude Sonnet 4.6 | Kimi K2.5 |
|---|---|---|
| LMArena Non-English | 1440 | 1433 |
| LMArena Chinese | 1491 | 1495 |
| LMArena French | 1465 | 1454 |
| LMArena German | 1428 | 1441 |
| LMArena Japanese | 1420 | 1421 |
| LMArena Korean | 1411 | 1410 |
| LMArena Russian | 1440 | 1435 |
| LMArena Spanish | 1464 | 1450 |
Instruction Following Claude Sonnet 4.6 leads
Claude Sonnet 4.6: 77.4 (#25), Kimi K2.5: 75.3 (#64)
| Benchmark | Claude Sonnet 4.6 | Kimi K2.5 |
|---|---|---|
| LMArena Instruction Following | 1475 | 1431 |
Long Context Kimi K2.5 leads
Claude Sonnet 4.6: 45.3 (#44), Kimi K2.5: 52.1 (#7)
| Benchmark | Claude Sonnet 4.6 | Kimi K2.5 |
|---|---|---|
| LMArena Longer Query | 1479 | 1445 |
| Fiction.LiveBench | — | 86.1% |
| CL-bench | — | 19.3% |
| CL-bench Life | — | 13.2% |
Writing & Preference Claude Sonnet 4.6 leads
Claude Sonnet 4.6: 70.2 (#22), Kimi K2.5: 65.1 (#53)
| Benchmark | Claude Sonnet 4.6 | Kimi K2.5 |
|---|---|---|
| LMArena Text | 1458 | 1445 |
| LMArena Creative Writing | 1435 | 1423 |
| EQ-Bench Creative Writing | 1810 | 1579 |
| LMArena Multi-Turn | 1464 | 1444 |
| EQ-Bench 4 | 1207 | — |
Frequently asked questions
Is Claude Sonnet 4.6 better than Kimi K2.5?
Claude Sonnet 4.6 is the stronger model overall, scoring 50.3 to 48.1 on the Noometry Index. Kimi K2.5 costs 6.7× less per token, which makes it the better buy when Claude Sonnet 4.6's lead doesn't matter for your workload.
Which is cheaper, Claude Sonnet 4.6 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 4.6 lists at $3 and $15.
Is Claude Sonnet 4.6 or Kimi K2.5 better for coding?
Kimi K2.5 scores higher on coding benchmarks: 48.8 versus 46.3 in the Noometry coding category.
Which has the bigger context window?
Claude Sonnet 4.6 does, with 1M tokens against 262K.
How many benchmarks do Claude Sonnet 4.6 and Kimi K2.5 share?
41 benchmarks have published results for both models. Claude Sonnet 4.6 has 57 scored results on Noometry and Kimi K2.5 has 51.