Model comparison
Claude Opus 4 vs Kimi K2 (Jul 2025)
Claude Opus 4 is the stronger model overall, scoring 42.9 to 40.9 on the Noometry Index. Kimi K2 (Jul 2025) costs 30× less per token, which makes it the better buy when Claude Opus 4's lead doesn't matter for your workload.
Last verified . 38 shared benchmarks.
Summary
- They share 38 benchmarks with published results for both. Claude Opus 4 scores higher in 6 categories and Kimi K2 (Jul 2025) in 3 categories; 7 gaps are clear of the uncertainty.
- The widest gap is in knowledge, where Claude Opus 4 leads 43.9 to 36.9.
- The biggest single-benchmark swing is SimpleBench: 58.8% for Claude Opus 4 and 26.3% for Kimi K2 (Jul 2025).
- Kimi K2 (Jul 2025) is cheaper at $0.57 / $2.30 per million input/output tokens, against $15 / $75 for Claude Opus 4.
- Kimi K2 (Jul 2025) accepts more context: 262K tokens versus 200K.
- Kimi K2 (Jul 2025) has downloadable open weights; the other is API-only.
Side by side
| Claude Opus 4 | Kimi K2 (Jul 2025) | |
|---|---|---|
| Provider | Anthropic | Moonshot AI |
| Noometry Index | 42.9 | 40.9 |
| Released | 2025-05-22 | 2025-07-12 |
| Weights | Proprietary | Open |
| Context window | 200K | 262K |
| Max output | 32K | 262K |
| Input $ / M tokens | $15 | $0.57 |
| Output $ / M tokens | $75 | $2.30 |
| Results tracked | 56 | 42 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding Claude Opus 4 leads
Claude Opus 4: 47.8 (#55), Kimi K2 (Jul 2025): 43.1 (#83)
| Benchmark | Claude Opus 4 | Kimi K2 (Jul 2025) |
|---|---|---|
| SWE-bench Verified (bash only) | 67.6% | 63.4% |
| Aider Polyglot | 72% | 59.1% |
| GSO | 6.9% | 4.9% |
| WeirdML | 43.7% | 42.8% |
| LMArena Coding | 1442 | 1399 |
| SWE-bench Verified | 70.7% | — |
| ALE-Bench | — | 597.5 |
| AlgoTune | 1.33 | — |
Agentic & Tool Use Claude Opus 4 leads
Claude Opus 4: 34.5 (#39), Kimi K2 (Jul 2025): 32.2 (#59)
| Benchmark | Claude Opus 4 | Kimi K2 (Jul 2025) |
|---|---|---|
| METR Time Horizons | 63.9% | 59.2% |
| Terminal-Bench | — | 35.7% |
| Berkeley Function Calling Leaderboard | — | 59.1% |
| Cybench | 38% | — |
| DeepResearch Bench | 46.8% | — |
| LMArena Search | 1127 | — |
Reasoning Claude Opus 4 leads
Claude Opus 4: 27.4 (#104), Kimi K2 (Jul 2025): 22.7 (#173)
| Benchmark | Claude Opus 4 | Kimi K2 (Jul 2025) |
|---|---|---|
| SimpleBench | 58.8% | 26.3% |
| Kagi LLM Benchmark | 74.3% | 64.4% |
| LMArena Hard Prompts | 1399 | 1384 |
| Epoch Capabilities Index | 142.67 | 146.02 |
| ForecastBench | 61.1 | 60.2 |
| ARC-AGI-2 | 8.6% | — |
| ARC-AGI-1 | 35.7% | — |
| CritPt | 0.3% | — |
| EnigmaEval | 5.6% | — |
| DTBench | 81.6% | — |
| LMCA | 37.4% | — |
Math Too close to call
Claude Opus 4: 40.2 (#80), Kimi K2 (Jul 2025): 40.1 (#81)
| Benchmark | Claude Opus 4 | Kimi K2 (Jul 2025) |
|---|---|---|
| Omni-MATH | 61.6% | 65.4% |
| LMArena Math | 1391 | 1396 |
| FrontierMath (Feb 2025 set) | 4.5% | 21.4% |
| FrontierMath Tier 4 (v1) | 4.2% | 0% |
| OTIS Mock AIME 2024-2025 | 64.4% | — |
| MATH Level 5 | 85% | — |
Knowledge Claude Opus 4 leads
Claude Opus 4: 43.9 (#82), Kimi K2 (Jul 2025): 36.9 (#153)
| Benchmark | Claude Opus 4 | Kimi K2 (Jul 2025) |
|---|---|---|
| MMLU-Pro | 87.5% | 81.9% |
| Confabulations | 15.9% | 20.4% |
| Vectara Hallucination Rate | 12% | 17.9% |
| GPQA (HELM) | 70.8% | 65.3% |
| LMArena Expert | 1384 | 1363 |
| GPQA Diamond | 76.3% | — |
| Humanity's Last Exam | 10.7% | — |
Multimodal Not comparable
Claude Opus 4: 30.9 (#101), Kimi K2 (Jul 2025): —
| Benchmark | Claude Opus 4 | Kimi K2 (Jul 2025) |
|---|---|---|
| LMArena Vision | 1192 | — |
| GeoBench | 49% | — |
| VPCT | 38% | — |
Multilingual Too close to call
Claude Opus 4: 49.0 (#133), Kimi K2 (Jul 2025): 49.8 (#125)
| Benchmark | Claude Opus 4 | Kimi K2 (Jul 2025) |
|---|---|---|
| LMArena Non-English | 1362 | 1372 |
| LMArena Chinese | 1386 | 1416 |
| LMArena French | 1370 | 1379 |
| LMArena German | 1391 | 1386 |
| LMArena Japanese | 1332 | 1349 |
| LMArena Korean | 1318 | 1324 |
| LMArena Russian | 1392 | 1384 |
| LMArena Spanish | 1388 | 1383 |
Instruction Following Claude Opus 4 leads
Claude Opus 4: 77.3 (#26), Kimi K2 (Jul 2025): 71.3 (#151)
| Benchmark | Claude Opus 4 | Kimi K2 (Jul 2025) |
|---|---|---|
| IFEval | 91.8% | 85% |
| LMArena Instruction Following | 1406 | 1347 |
Long Context Kimi K2 (Jul 2025) leads
Claude Opus 4: 39.5 (#167), Kimi K2 (Jul 2025): 41.2 (#140)
| Benchmark | Claude Opus 4 | Kimi K2 (Jul 2025) |
|---|---|---|
| Fiction.LiveBench | 61.1% | 66.7% |
| LMArena Longer Query | 1422 | 1353 |
| CL-bench | — | 17.6% |
Writing & Preference Kimi K2 (Jul 2025) leads
Claude Opus 4: 61.4 (#85), Kimi K2 (Jul 2025): 62.5 (#76)
| Benchmark | Claude Opus 4 | Kimi K2 (Jul 2025) |
|---|---|---|
| LMArena Text | 1377 | 1380 |
| LMArena Creative Writing | 1386 | 1349 |
| Short-Story Creative Writing | 83.6% | 85.6% |
| EQ-Bench Creative Writing | 1580 | 1666 |
| WildBench | 85.2% | 86.2% |
| LMArena Multi-Turn | 1396 | 1371 |
Frequently asked questions
Is Claude Opus 4 better than Kimi K2 (Jul 2025)?
Claude Opus 4 is the stronger model overall, scoring 42.9 to 40.9 on the Noometry Index. Kimi K2 (Jul 2025) costs 30× less per token, which makes it the better buy when Claude Opus 4's lead doesn't matter for your workload.
Which is cheaper, Claude Opus 4 or Kimi K2 (Jul 2025)?
Kimi K2 (Jul 2025) is cheaper. It lists at $0.57 per million input tokens and $2.30 per million output tokens; Claude Opus 4 lists at $15 and $75.
Is Claude Opus 4 or Kimi K2 (Jul 2025) better for coding?
Claude Opus 4 scores higher on coding benchmarks: 47.8 versus 43.1 in the Noometry coding category.
Which has the bigger context window?
Kimi K2 (Jul 2025) does, with 262K tokens against 200K.
How many benchmarks do Claude Opus 4 and Kimi K2 (Jul 2025) share?
38 benchmarks have published results for both models. Claude Opus 4 has 56 scored results on Noometry and Kimi K2 (Jul 2025) has 42.