Model comparison
Kimi K3 vs o1
Kimi K3 is the stronger model overall, scoring 59.5 to 40.9 on the Noometry Index.
Last verified . 28 shared benchmarks.
Summary
- They share 28 benchmarks with published results for both. Kimi K3 scores higher in 9 categories and o1 in 1 category; 10 gaps are clear of the uncertainty.
- The widest gap is in math, where Kimi K3 leads 74.2 to 36.1.
- The biggest single-benchmark swing is ARC-AGI-1: 94.5% for Kimi K3 and 30.7% for o1.
- Kimi K3 is cheaper at $3 / $15 per million input/output tokens, against $15 / $60 for o1.
- Kimi K3 accepts more context: 1.05M tokens versus 200K.
- Kimi K3 has downloadable open weights; the other is API-only.
Side by side
| Kimi K3 | o1 | |
|---|---|---|
| Provider | Moonshot AI | OpenAI |
| Noometry Index | 59.5 | 40.9 |
| Released | 2026-07-16 | 2024-09-12 |
| Weights | Open | Proprietary |
| Context window | 1.05M | 200K |
| Max output | 1.05M | 100K |
| Input $ / M tokens | $3 | $15 |
| Output $ / M tokens | $15 | $60 |
| Results tracked | 53 | 52 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding Kimi K3 leads
Kimi K3: 61.0 (#10), o1: 46.1 (#70)
| Benchmark | Kimi K3 | o1 |
|---|---|---|
| WeirdML | 82.6% | 47.6% |
| LMArena Coding | 1508 | 1367 |
| DeepSWE | 68.5% | — |
| FrontierCode | 44.2% | — |
| Aider Polyglot | — | 61.7% |
| LMArena WebDev | 1654 | — |
| FrontierSWE | 25.9% | — |
| SciCode | 59.5% | — |
| LiveBench Coding | — | 69.7% |
| CadEval | — | 56% |
| ALE-Bench | 1,524 | — |
| HumanEval+ | — | 89% |
| MBPP+ | — | 80.2% |
Agentic & Tool Use Kimi K3 leads
Kimi K3: 41.8 (#20), o1: 24.6 (#117)
| Benchmark | Kimi K3 | o1 |
|---|---|---|
| APEX-Agents | 50.6% | — |
| τ²-bench Banking | 37.1% | — |
| Cybench | — | 10% |
| PostTrainBench | 32% | — |
| GBAEval | 48.3% | — |
| GDP.pdf | 19% | — |
| METR Time Horizons | — | 51.1% |
| Vending-Bench 2 | 5,165 | — |
Reasoning Kimi K3 leads
Kimi K3: 63.0 (#17), o1: 27.9 (#111)
| Benchmark | Kimi K3 | o1 |
|---|---|---|
| SimpleBench | 60.7% | 41.7% |
| ARC-AGI-1 | 94.5% | 30.7% |
| Chess Puzzles | 39% | 15% |
| LMArena Hard Prompts | 1496 | 1371 |
| DTBench | 91.2% | 74.7% |
| LMCA | 52.7% | 22.3% |
| Epoch Capabilities Index | 157.45 | 141.91 |
| ARC-AGI-2 | 60.4% | — |
| NYT Connections (extended) | 93.6% | — |
| CritPt | 23.4% | — |
| EnigmaEval | — | 5.7% |
| LiveBench Reasoning | — | 91.6% |
| Mystery Game Puzzles | 26% | — |
| LiveBench Data Analysis | — | 65.5% |
| Surface Evolver Bench | 95% | — |
| ForecastBench | 61.1 | — |
| LiveBench | — | 75.7% |
Math Kimi K3 leads
Kimi K3: 74.2 (#16), o1: 36.1 (#175)
| Benchmark | Kimi K3 | o1 |
|---|---|---|
| FrontierMath (Tiers 1-3) | 72.2% | 14.7% |
| OTIS Mock AIME 2024-2025 | 97.2% | 73.3% |
| LMArena Math | 1491 | 1388 |
| FrontierMath Tier 4 | 39% | — |
| MathArena Final-Answer Competitions | 87.8% | — |
| ProofBench | 87% | — |
| LiveBench Math | — | 80.3% |
| MATH Level 5 | — | 94.7% |
| FrontierMath (Feb 2025 set) | — | 9.3% |
Knowledge Kimi K3 leads
Kimi K3: 63.2 (#21), o1: 41.5 (#110)
| Benchmark | Kimi K3 | o1 |
|---|---|---|
| GPQA Diamond | 93.1% | 76.8% |
| SimpleQA Verified | 50.6% | 41.1% |
| LMArena Expert | 1521 | 1361 |
| Humanity's Last Exam | — | 8% |
| Confabulations | — | 11.7% |
Multimodal Kimi K3 leads
Kimi K3: 37.8 (#70), o1: 34.2 (#93)
| Benchmark | Kimi K3 | o1 |
|---|---|---|
| LMArena Vision | — | 1168 |
| GeoBench | — | 80% |
| VPCT | — | 37% |
| Blueprint-Bench 2 | 29.5% | — |
| Furniture Assembly | 34.2% | — |
| SpatialViz-Bench | — | 41.4% |
Multilingual Kimi K3 leads
Kimi K3: 56.3 (#21), o1: 48.6 (#142)
| Benchmark | Kimi K3 | o1 |
|---|---|---|
| LMArena Non-English | 1466 | 1358 |
| LMArena Chinese | 1529 | 1394 |
| LMArena French | 1491 | 1344 |
| LMArena German | 1488 | 1337 |
| LMArena Japanese | 1487 | 1346 |
| LMArena Korean | 1458 | 1396 |
| LMArena Russian | 1482 | 1356 |
| LMArena Spanish | 1472 | 1345 |
Instruction Following Kimi K3 leads
Kimi K3: 77.7 (#14), o1: 74.8 (#86)
| Benchmark | Kimi K3 | o1 |
|---|---|---|
| LMArena Instruction Following | 1483 | 1367 |
| LiveBench Instruction Following | — | 81.5% |
Long Context o1 leads
Kimi K3: 45.8 (#29), o1: 50.3 (#9)
| Benchmark | Kimi K3 | o1 |
|---|---|---|
| LMArena Longer Query | 1494 | 1378 |
| Fiction.LiveBench | — | 83.3% |
Writing & Preference Kimi K3 leads
Kimi K3: 76.6 (#4), o1: 55.6 (#144)
| Benchmark | Kimi K3 | o1 |
|---|---|---|
| LMArena Text | 1476 | 1366 |
| LMArena Creative Writing | 1454 | 1348 |
| LMArena Multi-Turn | 1488 | 1369 |
| Short-Story Creative Writing | — | 70.2% |
| EQ-Bench Creative Writing | 2082 | — |
| EQ-Bench 4 | 1339 | — |
| LiveBench Language | — | 65.4% |
Frequently asked questions
Is Kimi K3 better than o1?
Kimi K3 is the stronger model overall, scoring 59.5 to 40.9 on the Noometry Index.
Which is cheaper, Kimi K3 or o1?
Kimi K3 is cheaper. It lists at $3 per million input tokens and $15 per million output tokens; o1 lists at $15 and $60.
Is Kimi K3 or o1 better for coding?
Kimi K3 scores higher on coding benchmarks: 61.0 versus 46.1 in the Noometry coding category.
Which has the bigger context window?
Kimi K3 does, with 1.05M tokens against 200K.
How many benchmarks do Kimi K3 and o1 share?
28 benchmarks have published results for both models. Kimi K3 has 53 scored results on Noometry and o1 has 52.