Model comparison
Claude Sonnet 5 vs Kimi K2.5
Claude Sonnet 5 is the stronger model overall, scoring 54.6 to 48.1 on the Noometry Index. Kimi K2.5 costs 4.4× less per token, which makes it the better buy when Claude Sonnet 5's lead doesn't matter for your workload.
Last verified . 33 shared benchmarks.
Summary
- They share 33 benchmarks with published results for both. Claude Sonnet 5 scores higher in 8 categories and Kimi K2.5 in 2 categories; 8 gaps are clear of the uncertainty.
- The widest gap is in reasoning, where Claude Sonnet 5 leads 49.1 to 31.2.
- The biggest single-benchmark swing is WeirdML: 68.8% for Claude Sonnet 5 and 45.6% for Kimi K2.5.
- Kimi K2.5 is cheaper at $0.45 / $2.25 per million input/output tokens, against $2 / $10 for Claude Sonnet 5.
- Claude Sonnet 5 accepts more context: 1M tokens versus 262K.
- Kimi K2.5 has downloadable open weights; the other is API-only.
Side by side
| Claude Sonnet 5 | Kimi K2.5 | |
|---|---|---|
| Provider | Anthropic | Moonshot AI |
| Noometry Index | 54.6 | 48.1 |
| Released | 2026-06-29 | 2026-01-27 |
| Weights | Proprietary | Open |
| Context window | 1M | 262K |
| Max output | 128K | 262K |
| Input $ / M tokens | $2 | $0.45 |
| Output $ / M tokens | $10 | $2.25 |
| 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 K2.5: 48.8 (#53)
| Benchmark | Claude Sonnet 5 | Kimi K2.5 |
|---|---|---|
| LMArena WebDev | 1541 | 1437 |
| SciCode | 54.3% | 49% |
| WeirdML | 68.8% | 45.6% |
| LMArena Coding | 1483 | 1474 |
| ALE-Bench | 1,463 | 821.65 |
| SWE-bench Verified | — | 73.8% |
| DeepSWE | 53.8% | — |
| FrontierCode | 42.7% | — |
| SWE-bench Verified (bash only) | — | 70.8% |
| CursorBench | 34.1% | — |
| SWE-bench Multilingual | — | 67.3% |
| GSO | 37.3% | — |
Agentic & Tool Use Claude Sonnet 5 leads
Claude Sonnet 5: 42.8 (#18), Kimi K2.5: 34.2 (#48)
| Benchmark | Claude Sonnet 5 | Kimi K2.5 |
|---|---|---|
| Vending-Bench 2 | 6,378 | 1,198 |
| Terminal-Bench | — | 43.2% |
| APEX-Agents | 54.5% | — |
| OSWorld | — | 63.3% |
| GBAEval | 65.3% | — |
| LMArena Search | 1194 | — |
Reasoning Claude Sonnet 5 leads
Claude Sonnet 5: 49.1 (#39), Kimi K2.5: 31.2 (#80)
| Benchmark | Claude Sonnet 5 | Kimi K2.5 |
|---|---|---|
| SimpleBench | 60.6% | 46.8% |
| NYT Connections (extended) | 75.1% | 69.9% |
| CritPt | 16.9% | 3.1% |
| Chess Puzzles | 35% | 12% |
| LMArena Hard Prompts | 1461 | 1453 |
| Epoch Capabilities Index | 156.21 | 148.03 |
| ARC-AGI-2 | — | 11.8% |
| Kagi LLM Benchmark | — | 78.5% |
| ARC-AGI-1 | — | 65.3% |
| EnigmaEval | — | 3.4% |
| Thematic Generalization | — | 69.4% |
| Mystery Game Puzzles | 35% | — |
| DTBench | 92.5% | — |
| LMCA | 50% | — |
| 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 K2.5: 51.8 (#53)
| Benchmark | Claude Sonnet 5 | Kimi K2.5 |
|---|---|---|
| OTIS Mock AIME 2024-2025 | 94.7% | 92.2% |
| LMArena Math | 1467 | 1470 |
| FrontierMath (Tiers 1-3) | 65.6% | — |
| FrontierMath Tier 4 | 29.3% | — |
| MathArena Final-Answer Competitions | — | 62.3% |
| ProofBench | 77% | — |
| FrontierMath (Feb 2025 set) | — | 27.9% |
| FrontierMath Tier 4 (v1) | — | 4.2% |
Knowledge Claude Sonnet 5 leads
Claude Sonnet 5: 55.6 (#47), Kimi K2.5: 53.6 (#56)
| Benchmark | Claude Sonnet 5 | Kimi K2.5 |
|---|---|---|
| GPQA Diamond | 90.5% | 87.6% |
| SimpleQA Verified | 33.7% | 34.3% |
| LMArena Expert | 1490 | 1466 |
| Humanity's Last Exam | — | 24.4% |
| Vectara Hallucination Rate | — | 14.2% |
Multimodal Claude Sonnet 5 leads
Claude Sonnet 5: 42.4 (#31), Kimi K2.5: 41.1 (#39)
| Benchmark | Claude Sonnet 5 | Kimi K2.5 |
|---|---|---|
| LMArena Vision | 1274 | 1269 |
| LMArena Document | 1466 | 1430 |
| Blueprint-Bench 2 | 24.9% | — |
Multilingual Too close to call
Claude Sonnet 5: 53.8 (#55), Kimi K2.5: 53.9 (#53)
| Benchmark | Claude Sonnet 5 | Kimi K2.5 |
|---|---|---|
| LMArena Non-English | 1431 | 1433 |
| LMArena Chinese | 1477 | 1495 |
| LMArena French | 1460 | 1454 |
| LMArena German | 1440 | 1441 |
| LMArena Japanese | 1422 | 1421 |
| LMArena Korean | 1411 | 1410 |
| LMArena Russian | 1451 | 1435 |
| LMArena Spanish | 1437 | 1450 |
Instruction Following Too close to call
Claude Sonnet 5: 76.3 (#41), Kimi K2.5: 75.3 (#64)
| Benchmark | Claude Sonnet 5 | Kimi K2.5 |
|---|---|---|
| LMArena Instruction Following | 1452 | 1431 |
Long Context Kimi K2.5 leads
Claude Sonnet 5: 44.8 (#55), Kimi K2.5: 52.1 (#7)
| Benchmark | Claude Sonnet 5 | Kimi K2.5 |
|---|---|---|
| LMArena Longer Query | 1463 | 1445 |
| Fiction.LiveBench | — | 86.1% |
| CL-bench | — | 19.3% |
| CL-bench Life | — | 13.2% |
Writing & Preference Claude Sonnet 5 leads
Claude Sonnet 5: 69.2 (#25), Kimi K2.5: 65.1 (#53)
| Benchmark | Claude Sonnet 5 | Kimi K2.5 |
|---|---|---|
| LMArena Text | 1442 | 1445 |
| LMArena Creative Writing | 1416 | 1423 |
| EQ-Bench Creative Writing | 1794 | 1579 |
| LMArena Multi-Turn | 1454 | 1444 |
| EQ-Bench 4 | 1236 | — |
Frequently asked questions
Is Claude Sonnet 5 better than Kimi K2.5?
Claude Sonnet 5 is the stronger model overall, scoring 54.6 to 48.1 on the Noometry Index. Kimi K2.5 costs 4.4× 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 K2.5?
Kimi K2.5 is cheaper. It lists at $0.45 per million input tokens and $2.25 per million output tokens; Claude Sonnet 5 lists at $2 and $10.
Is Claude Sonnet 5 or Kimi K2.5 better for coding?
Claude Sonnet 5 scores higher on coding benchmarks: 55.5 versus 48.8 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 K2.5 share?
33 benchmarks have published results for both models. Claude Sonnet 5 has 51 scored results on Noometry and Kimi K2.5 has 51.