Model comparison
Kimi K2 (Jul 2025) vs o3-mini
Kimi K2 (Jul 2025) is the stronger model overall, scoring 41.2 to 36.7 on the Noometry Index.
Last verified . 28 shared benchmarks.
Summary
- They share 28 benchmarks with published results for both. Kimi K2 (Jul 2025) scores higher in 7 categories and o3-mini in 2 categories; 8 gaps are clear of the uncertainty.
- The widest gap is in math, where Kimi K2 (Jul 2025) leads 42.7 to 28.1.
- The biggest single-benchmark swing is Short-Story Creative Writing: 85.6% for Kimi K2 (Jul 2025) and 61.7% for o3-mini.
- Kimi K2 (Jul 2025) is cheaper at $0.57 / $2.30 per million input/output tokens, against $1.10 / $4.40 for o3-mini.
- 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
| Kimi K2 (Jul 2025) | o3-mini | |
|---|---|---|
| Provider | Moonshot AI | OpenAI |
| Noometry Index | 41.2 | 36.7 |
| Released | 2025-07-12 | 2024-12-20 |
| Weights | Open | Proprietary |
| Context window | 262K | 200K |
| Max output | 262K | 100K |
| Input $ / M tokens | $0.57 | $1.10 |
| Output $ / M tokens | $2.30 | $4.40 |
| Results tracked | 42 | 51 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding Kimi K2 (Jul 2025) leads
Kimi K2 (Jul 2025): 42.4 (#102), o3-mini: 40.8 (#132)
| Benchmark | Kimi K2 (Jul 2025) | o3-mini |
|---|---|---|
| Aider Polyglot | 59.1% | 60.4% |
| GSO | 4.9% | 1.3% |
| WeirdML | 42.8% | 43.7% |
| LMArena Coding | 1399 | 1378 |
| SWE-bench Verified (bash only) | 63.4% | — |
| SciCode | — | 39.8% |
| LiveBench Coding | — | 82.7% |
| CadEval | — | 54% |
| ALE-Bench | 597.5 | — |
Agentic & Tool Use Kimi K2 (Jul 2025) leads
Kimi K2 (Jul 2025): 32.4 (#64), o3-mini: 29.6 (#84)
| Benchmark | Kimi K2 (Jul 2025) | o3-mini |
|---|---|---|
| Terminal-Bench | 35.7% | — |
| Berkeley Function Calling Leaderboard | 59.1% | — |
| Cybench | — | 22.5% |
| METR Time Horizons | 59.2% | — |
Reasoning Kimi K2 (Jul 2025) leads
Kimi K2 (Jul 2025): 23.3 (#179), o3-mini: 16.3 (#305)
| Benchmark | Kimi K2 (Jul 2025) | o3-mini |
|---|---|---|
| SimpleBench | 26.3% | 22.8% |
| LMArena Hard Prompts | 1384 | 1366 |
| Epoch Capabilities Index | 146.01 | 140.34 |
| ForecastBench | 60.2 | 59.6 |
| ARC-AGI-2 | — | 3% |
| Kagi LLM Benchmark | 64.4% | — |
| ARC-AGI-1 | — | 34.5% |
| CritPt | — | 0.3% |
| Chess Puzzles | — | 17% |
| LiveBench Reasoning | — | 89.6% |
| Mystery Game Puzzles | — | 7% |
| DTBench | — | 68.8% |
| LiveBench Data Analysis | — | 70.6% |
| LMCA | — | 19% |
| LiveBench | — | 75.9% |
Math Kimi K2 (Jul 2025) leads
Kimi K2 (Jul 2025): 42.7 (#83), o3-mini: 28.1 (#244)
| Benchmark | Kimi K2 (Jul 2025) | o3-mini |
|---|---|---|
| LMArena Math | 1397 | 1396 |
| FrontierMath (Feb 2025 set) | 21.4% | 12.4% |
| FrontierMath Tier 4 (v1) | 0% | 4.2% |
| FrontierMath (Tiers 1-3) | — | 18.6% |
| FrontierMath Tier 4 | — | 0% |
| OTIS Mock AIME 2024-2025 | — | 76.9% |
| Omni-MATH | 65.4% | — |
| LiveBench Math | — | 77.3% |
| MATH Level 5 | — | 96.5% |
Knowledge Too close to call
Kimi K2 (Jul 2025): 37.3 (#157), o3-mini: 38.3 (#146)
| Benchmark | Kimi K2 (Jul 2025) | o3-mini |
|---|---|---|
| Confabulations | 20.4% | 17.9% |
| LMArena Expert | 1365 | 1364 |
| GPQA Diamond | — | 77% |
| SimpleQA Verified | — | 15.3% |
| MMLU-Pro | 81.9% | — |
| Vectara Hallucination Rate | 17.9% | — |
| GPQA (HELM) | 65.3% | — |
Multilingual Kimi K2 (Jul 2025) leads
Kimi K2 (Jul 2025): 49.6 (#130), o3-mini: 45.7 (#164)
| Benchmark | Kimi K2 (Jul 2025) | o3-mini |
|---|---|---|
| LMArena Non-English | 1372 | 1319 |
| LMArena Chinese | 1415 | 1379 |
| LMArena French | 1379 | 1334 |
| LMArena German | 1387 | 1303 |
| LMArena Japanese | 1349 | 1286 |
| LMArena Korean | 1325 | 1314 |
| LMArena Russian | 1385 | 1304 |
| LMArena Spanish | 1386 | 1321 |
Instruction Following o3-mini leads
Kimi K2 (Jul 2025): 71.1 (#156), o3-mini: 75.1 (#72)
| Benchmark | Kimi K2 (Jul 2025) | o3-mini |
|---|---|---|
| LMArena Instruction Following | 1348 | 1337 |
| LiveBench Instruction Following | — | 84.4% |
| IFEval | 85% | — |
Long Context Kimi K2 (Jul 2025) leads
Kimi K2 (Jul 2025): 41.2 (#145), o3-mini: 33.8 (#256)
| Benchmark | Kimi K2 (Jul 2025) | o3-mini |
|---|---|---|
| Fiction.LiveBench | 66.7% | 50% |
| LMArena Longer Query | 1353 | 1343 |
| CL-bench | 17.6% | — |
Writing & Preference Kimi K2 (Jul 2025) leads
Kimi K2 (Jul 2025): 62.3 (#78), o3-mini: 50.3 (#182)
| Benchmark | Kimi K2 (Jul 2025) | o3-mini |
|---|---|---|
| LMArena Text | 1380 | 1337 |
| LMArena Creative Writing | 1350 | 1286 |
| Short-Story Creative Writing | 85.6% | 61.7% |
| LMArena Multi-Turn | 1371 | 1320 |
| EQ-Bench Creative Writing | 1666 | — |
| WildBench | 86.2% | — |
| LiveBench Language | — | 50.7% |
Frequently asked questions
Is Kimi K2 (Jul 2025) better than o3-mini?
Kimi K2 (Jul 2025) is the stronger model overall, scoring 41.2 to 36.7 on the Noometry Index.
Which is cheaper, Kimi K2 (Jul 2025) or o3-mini?
Kimi K2 (Jul 2025) is cheaper. It lists at $0.57 per million input tokens and $2.30 per million output tokens; o3-mini lists at $1.10 and $4.40.
Is Kimi K2 (Jul 2025) or o3-mini better for coding?
Kimi K2 (Jul 2025) scores higher on coding benchmarks: 42.4 versus 40.8 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 Kimi K2 (Jul 2025) and o3-mini share?
28 benchmarks have published results for both models. Kimi K2 (Jul 2025) has 42 scored results on Noometry and o3-mini has 51.