Model comparison
Claude Opus 4 vs Kimi K3
Kimi K3 is the stronger model overall, scoring 59.5 to 43.1 on the Noometry Index.
Last verified . 29 shared benchmarks.
Summary
- They share 29 benchmarks with published results for both. Claude Opus 4 scores higher in 0 categories and Kimi K3 in 10 categories; 9 gaps are clear of the uncertainty.
- The widest gap is in reasoning, where Kimi K3 leads 63.0 to 27.3.
- The biggest single-benchmark swing is ARC-AGI-1: 35.7% for Claude Opus 4 and 94.5% for Kimi K3.
- Kimi K3 is cheaper at $3 / $15 per million input/output tokens, against $15 / $75 for Claude Opus 4.
- Kimi K3 accepts more context: 1.05M tokens versus 200K.
- Kimi K3 has downloadable open weights; the other is API-only.
Side by side
| Claude Opus 4 | Kimi K3 | |
|---|---|---|
| Provider | Anthropic | Moonshot AI |
| Noometry Index | 43.1 | 59.5 |
| Released | 2025-05-22 | 2026-07-16 |
| Weights | Proprietary | Open |
| Context window | 200K | 1.05M |
| Max output | 32K | 1.05M |
| Input $ / M tokens | $15 | $3 |
| Output $ / M tokens | $75 | $15 |
| Results tracked | 56 | 53 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding Kimi K3 leads
Claude Opus 4: 47.2 (#62), Kimi K3: 61.0 (#10)
| Benchmark | Claude Opus 4 | Kimi K3 |
|---|---|---|
| WeirdML | 43.7% | 82.6% |
| LMArena Coding | 1442 | 1508 |
| SWE-bench Verified | 70.7% | — |
| DeepSWE | — | 68.5% |
| FrontierCode | — | 44.2% |
| SWE-bench Verified (bash only) | 67.6% | — |
| Aider Polyglot | 72% | — |
| LMArena WebDev | — | 1654 |
| FrontierSWE | — | 25.9% |
| SciCode | — | 59.5% |
| GSO | 6.9% | — |
| ALE-Bench | — | 1,524 |
| AlgoTune | 1.33 | — |
Agentic & Tool Use Kimi K3 leads
Claude Opus 4: 34.8 (#42), Kimi K3: 41.8 (#20)
| Benchmark | Claude Opus 4 | Kimi K3 |
|---|---|---|
| APEX-Agents | — | 50.6% |
| τ²-bench Banking | — | 37.1% |
| Cybench | 38% | — |
| DeepResearch Bench | 46.8% | — |
| PostTrainBench | — | 32% |
| GBAEval | — | 48.3% |
| GDP.pdf | — | 19% |
| LMArena Search | 1127 | — |
| METR Time Horizons | 63.9% | — |
| Vending-Bench 2 | — | 5,165 |
Reasoning Kimi K3 leads
Claude Opus 4: 27.3 (#121), Kimi K3: 63.0 (#17)
| Benchmark | Claude Opus 4 | Kimi K3 |
|---|---|---|
| ARC-AGI-2 | 8.6% | 60.4% |
| SimpleBench | 58.8% | 60.7% |
| ARC-AGI-1 | 35.7% | 94.5% |
| CritPt | 0.3% | 23.4% |
| LMArena Hard Prompts | 1399 | 1496 |
| DTBench | 81.6% | 91.2% |
| LMCA | 37.4% | 52.7% |
| Epoch Capabilities Index | 142.67 | 157.45 |
| ForecastBench | 61.1 | 61.1 |
| Kagi LLM Benchmark | 74.3% | — |
| NYT Connections (extended) | — | 93.6% |
| Chess Puzzles | — | 39% |
| EnigmaEval | 5.6% | — |
| Mystery Game Puzzles | — | 26% |
| Surface Evolver Bench | — | 95% |
Math Kimi K3 leads
Claude Opus 4: 42.0 (#86), Kimi K3: 74.2 (#16)
| Benchmark | Claude Opus 4 | Kimi K3 |
|---|---|---|
| OTIS Mock AIME 2024-2025 | 64.4% | 97.2% |
| LMArena Math | 1390 | 1491 |
| FrontierMath (Tiers 1-3) | — | 72.2% |
| FrontierMath Tier 4 | — | 39% |
| MathArena Final-Answer Competitions | — | 87.8% |
| ProofBench | — | 87% |
| Omni-MATH | 61.6% | — |
| MATH Level 5 | 85% | — |
| FrontierMath (Feb 2025 set) | 4.5% | — |
| FrontierMath Tier 4 (v1) | 4.2% | — |
Knowledge Kimi K3 leads
Claude Opus 4: 44.0 (#88), Kimi K3: 63.2 (#21)
| Benchmark | Claude Opus 4 | Kimi K3 |
|---|---|---|
| GPQA Diamond | 76.3% | 93.1% |
| LMArena Expert | 1386 | 1521 |
| Humanity's Last Exam | 10.7% | — |
| SimpleQA Verified | — | 50.6% |
| MMLU-Pro | 87.5% | — |
| Confabulations | 15.9% | — |
| Vectara Hallucination Rate | 12% | — |
| GPQA (HELM) | 70.8% | — |
Multimodal Kimi K3 leads
Claude Opus 4: 31.5 (#106), Kimi K3: 37.8 (#70)
| Benchmark | Claude Opus 4 | Kimi K3 |
|---|---|---|
| LMArena Vision | 1192 | — |
| GeoBench | 49% | — |
| VPCT | 38% | — |
| Blueprint-Bench 2 | — | 29.5% |
| Furniture Assembly | — | 34.2% |
Multilingual Kimi K3 leads
Claude Opus 4: 48.8 (#138), Kimi K3: 56.3 (#21)
| Benchmark | Claude Opus 4 | Kimi K3 |
|---|---|---|
| LMArena Non-English | 1362 | 1466 |
| LMArena Chinese | 1386 | 1529 |
| LMArena French | 1372 | 1491 |
| LMArena German | 1391 | 1488 |
| LMArena Japanese | 1331 | 1487 |
| LMArena Korean | 1321 | 1458 |
| LMArena Russian | 1392 | 1482 |
| LMArena Spanish | 1389 | 1472 |
Instruction Following Too close to call
Claude Opus 4: 77.1 (#28), Kimi K3: 77.7 (#14)
| Benchmark | Claude Opus 4 | Kimi K3 |
|---|---|---|
| LMArena Instruction Following | 1406 | 1483 |
| IFEval | 91.8% | — |
Long Context Kimi K3 leads
Claude Opus 4: 39.6 (#172), Kimi K3: 45.8 (#29)
| Benchmark | Claude Opus 4 | Kimi K3 |
|---|---|---|
| LMArena Longer Query | 1422 | 1494 |
| Fiction.LiveBench | 61.1% | — |
Writing & Preference Kimi K3 leads
Claude Opus 4: 61.2 (#89), Kimi K3: 76.6 (#4)
| Benchmark | Claude Opus 4 | Kimi K3 |
|---|---|---|
| LMArena Text | 1377 | 1476 |
| LMArena Creative Writing | 1387 | 1454 |
| EQ-Bench Creative Writing | 1580 | 2082 |
| LMArena Multi-Turn | 1396 | 1488 |
| Short-Story Creative Writing | 83.6% | — |
| WildBench | 85.2% | — |
| EQ-Bench 4 | — | 1339 |
Frequently asked questions
Is Claude Opus 4 better than Kimi K3?
Kimi K3 is the stronger model overall, scoring 59.5 to 43.1 on the Noometry Index.
Which is cheaper, Claude Opus 4 or Kimi K3?
Kimi K3 is cheaper. It lists at $3 per million input tokens and $15 per million output tokens; Claude Opus 4 lists at $15 and $75.
Is Claude Opus 4 or Kimi K3 better for coding?
Kimi K3 scores higher on coding benchmarks: 61.0 versus 47.2 in the Noometry coding category.
Which has the bigger context window?
Kimi K3 does, with 1.05M tokens against 200K.
How many benchmarks do Claude Opus 4 and Kimi K3 share?
29 benchmarks have published results for both models. Claude Opus 4 has 56 scored results on Noometry and Kimi K3 has 53.