Model comparison
Claude Sonnet 5 vs Kimi K2.6
Claude Sonnet 5 is the stronger model overall, scoring 54.6 to 47.7 on the Noometry Index. Kimi K2.6 costs 2.3× less per token, which makes it the better buy when Claude Sonnet 5'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 5 scores higher in 8 categories and Kimi K2.6 in 2 categories; 7 gaps are clear of the uncertainty.
- The widest gap is in agentic & tool use, where Claude Sonnet 5 leads 42.8 to 21.9.
- The biggest single-benchmark swing is GBAEval: 65.3% for Claude Sonnet 5 and 0.9% for Kimi K2.6.
- Kimi K2.6 is cheaper at $0.95 / $4 per million input/output tokens, against $2 / $10 for Claude Sonnet 5.
- Claude Sonnet 5 accepts more context: 1M tokens versus 262K.
- Kimi K2.6 has downloadable open weights; the other is API-only.
Side by side
| Claude Sonnet 5 | Kimi K2.6 | |
|---|---|---|
| Provider | Anthropic | Moonshot AI |
| Noometry Index | 54.6 | 47.7 |
| Released | 2026-06-29 | 2026-04-20 |
| Weights | Proprietary | Open |
| Context window | 1M | 262K |
| Max output | 128K | 262K |
| Input $ / M tokens | $2 | $0.95 |
| Output $ / M tokens | $10 | $4 |
| Results tracked | 51 | 51 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding Claude Sonnet 5 leads
Claude Sonnet 5: 55.5 (#26), Kimi K2.6: 50.7 (#43)
| Benchmark | Claude Sonnet 5 | Kimi K2.6 |
|---|---|---|
| LMArena WebDev | 1541 | 1509 |
| SciCode | 54.3% | 53.5% |
| WeirdML | 68.8% | 55.9% |
| LMArena Coding | 1483 | 1488 |
| ALE-Bench | 1,463 | 1,093 |
| SWE-bench Verified | — | 76.7% |
| DeepSWE | 53.8% | — |
| FrontierCode | 42.7% | — |
| CursorBench | 34.1% | — |
| GSO | 37.3% | — |
Agentic & Tool Use Claude Sonnet 5 leads
Claude Sonnet 5: 42.8 (#18), Kimi K2.6: 21.9 (#137)
| Benchmark | Claude Sonnet 5 | Kimi K2.6 |
|---|---|---|
| GBAEval | 65.3% | 0.9% |
| Vending-Bench 2 | 6,378 | 6,205 |
| APEX-Agents | 54.5% | — |
| OSWorld 2.0 | — | 4.6% |
| ExploitBench | — | 18.4% |
| GDP.pdf | — | 12% |
| LMArena Search | 1194 | — |
Reasoning Claude Sonnet 5 leads
Claude Sonnet 5: 49.1 (#39), Kimi K2.6: 40.5 (#55)
| Benchmark | Claude Sonnet 5 | Kimi K2.6 |
|---|---|---|
| NYT Connections (extended) | 75.1% | 87.2% |
| CritPt | 16.9% | 8% |
| Chess Puzzles | 35% | 26% |
| LMArena Hard Prompts | 1461 | 1470 |
| Mystery Game Puzzles | 35% | 18% |
| DTBench | 92.5% | 90.9% |
| LMCA | 50% | 37.3% |
| Epoch Capabilities Index | 156.21 | 151.05 |
| SimpleBench | 60.6% | — |
| EBR-Bench | — | 2.4% |
| Surface Evolver Bench | 60% | — |
| Bench to the Future 3 | 0.14 | — |
| ForecastBench | 61.1 | — |
Math Claude Sonnet 5 leads
Claude Sonnet 5: 66.2 (#27), Kimi K2.6: 57.0 (#41)
| Benchmark | Claude Sonnet 5 | Kimi K2.6 |
|---|---|---|
| FrontierMath (Tiers 1-3) | 65.6% | 57.2% |
| FrontierMath Tier 4 | 29.3% | 25.6% |
| OTIS Mock AIME 2024-2025 | 94.7% | 96.1% |
| ProofBench | 77% | 16% |
| LMArena Math | 1467 | 1475 |
| MathArena Final-Answer Competitions | — | 72.9% |
| FrontierMath (Feb 2025 set) | — | 39% |
| FrontierMath Tier 4 (v1) | — | 14.6% |
Knowledge Claude Sonnet 5 leads
Claude Sonnet 5: 55.6 (#47), Kimi K2.6: 54.0 (#54)
| Benchmark | Claude Sonnet 5 | Kimi K2.6 |
|---|---|---|
| GPQA Diamond | 90.5% | 90.8% |
| SimpleQA Verified | 33.7% | 34.9% |
| LMArena Expert | 1490 | 1491 |
| Vectara Hallucination Rate | — | 10.8% |
Multimodal Claude Sonnet 5 leads
Claude Sonnet 5: 42.4 (#31), Kimi K2.6: 31.6 (#103)
| Benchmark | Claude Sonnet 5 | Kimi K2.6 |
|---|---|---|
| LMArena Vision | 1274 | 1283 |
| Blueprint-Bench 2 | 24.9% | 3.9% |
| LMArena Document | 1466 | 1451 |
| Furniture Assembly | — | 21.7% |
Multilingual Kimi K2.6 leads
Claude Sonnet 5: 53.8 (#55), Kimi K2.6: 54.9 (#37)
| Benchmark | Claude Sonnet 5 | Kimi K2.6 |
|---|---|---|
| LMArena Non-English | 1431 | 1446 |
| LMArena Chinese | 1477 | 1521 |
| LMArena French | 1460 | 1471 |
| LMArena German | 1440 | 1450 |
| LMArena Japanese | 1422 | 1443 |
| LMArena Korean | 1411 | 1427 |
| LMArena Russian | 1451 | 1446 |
| LMArena Spanish | 1437 | 1464 |
Instruction Following Too close to call
Claude Sonnet 5: 76.3 (#41), Kimi K2.6: 76.3 (#43)
| Benchmark | Claude Sonnet 5 | Kimi K2.6 |
|---|---|---|
| LMArena Instruction Following | 1452 | 1451 |
Long Context Too close to call
Claude Sonnet 5: 44.8 (#55), Kimi K2.6: 44.9 (#52)
| Benchmark | Claude Sonnet 5 | Kimi K2.6 |
|---|---|---|
| LMArena Longer Query | 1463 | 1468 |
Writing & Preference Too close to call
Claude Sonnet 5: 69.2 (#25), Kimi K2.6: 68.5 (#26)
| Benchmark | Claude Sonnet 5 | Kimi K2.6 |
|---|---|---|
| LMArena Text | 1442 | 1455 |
| LMArena Creative Writing | 1416 | 1434 |
| EQ-Bench Creative Writing | 1794 | 1725 |
| EQ-Bench 4 | 1236 | 1202 |
| LMArena Multi-Turn | 1454 | 1453 |
Frequently asked questions
Is Claude Sonnet 5 better than Kimi K2.6?
Claude Sonnet 5 is the stronger model overall, scoring 54.6 to 47.7 on the Noometry Index. Kimi K2.6 costs 2.3× less per token, which makes it the better buy when Claude Sonnet 5's lead doesn't matter for your workload.
Which is cheaper, Claude Sonnet 5 or Kimi K2.6?
Kimi K2.6 is cheaper. It lists at $0.95 per million input tokens and $4 per million output tokens; Claude Sonnet 5 lists at $2 and $10.
Is Claude Sonnet 5 or Kimi K2.6 better for coding?
Claude Sonnet 5 scores higher on coding benchmarks: 55.5 versus 50.7 in the Noometry coding category.
Which has the bigger context window?
Claude Sonnet 5 does, with 1M tokens against 262K.
How many benchmarks do Claude Sonnet 5 and Kimi K2.6 share?
41 benchmarks have published results for both models. Claude Sonnet 5 has 51 scored results on Noometry and Kimi K2.6 has 51.