Model comparison
Kimi K3 vs o4-mini
Kimi K3 is the stronger model overall, scoring 59.5 to 41.6 on the Noometry Index. o4-mini costs 3.1× less per token, which makes it the better buy when Kimi K3's lead doesn't matter for your workload.
Last verified . 34 shared benchmarks.
Summary
- They share 34 benchmarks with published results for both. Kimi K3 scores higher in 9 categories and o4-mini in 1 category; 9 gaps are clear of the uncertainty.
- The widest gap is in reasoning, where Kimi K3 leads 63.0 to 24.6.
- The biggest single-benchmark swing is ARC-AGI-2: 60.4% for Kimi K3 and 6.1% for o4-mini.
- o4-mini is cheaper at $1.10 / $4.40 per million input/output tokens, against $3 / $15 for Kimi K3.
- 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 | o4-mini | |
|---|---|---|
| Provider | Moonshot AI | OpenAI |
| Noometry Index | 59.5 | 41.6 |
| Released | 2026-07-16 | 2025-04-16 |
| Weights | Open | Proprietary |
| Context window | 1.05M | 200K |
| Max output | 1.05M | 100K |
| Input $ / M tokens | $3 | $1.10 |
| Output $ / M tokens | $15 | $4.40 |
| Results tracked | 53 | 60 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding Kimi K3 leads
Kimi K3: 61.0 (#10), o4-mini: 40.9 (#127)
| Benchmark | Kimi K3 | o4-mini |
|---|---|---|
| WeirdML | 82.6% | 52.6% |
| LMArena Coding | 1508 | 1368 |
| ALE-Bench | 1,524 | 826.17 |
| DeepSWE | 68.5% | — |
| FrontierCode | 44.2% | — |
| SWE-bench Verified (bash only) | — | 45% |
| Aider Polyglot | — | 72% |
| LMArena WebDev | 1654 | — |
| FrontierSWE | 25.9% | — |
| SciCode | 59.5% | — |
| GSO | — | 3.6% |
| CadEval | — | 62% |
| AlgoTune | — | 1.72 |
Agentic & Tool Use Kimi K3 leads
Kimi K3: 41.8 (#20), o4-mini: 32.6 (#61)
| Benchmark | Kimi K3 | o4-mini |
|---|---|---|
| APEX-Agents | 50.6% | — |
| Berkeley Function Calling Leaderboard | — | 53.2% |
| GDPval | — | 25.3% |
| τ²-bench Banking | 37.1% | — |
| PostTrainBench | 32% | — |
| GBAEval | 48.3% | — |
| GDP.pdf | 19% | — |
| METR Time Horizons | — | 63.9% |
| Vending-Bench 2 | 5,165 | — |
Reasoning Kimi K3 leads
Kimi K3: 63.0 (#17), o4-mini: 24.6 (#162)
| Benchmark | Kimi K3 | o4-mini |
|---|---|---|
| ARC-AGI-2 | 60.4% | 6.1% |
| SimpleBench | 60.7% | 38.7% |
| ARC-AGI-1 | 94.5% | 58.7% |
| CritPt | 23.4% | 0.6% |
| Chess Puzzles | 39% | 26% |
| LMArena Hard Prompts | 1496 | 1351 |
| Mystery Game Puzzles | 26% | 5% |
| DTBench | 91.2% | 77.6% |
| LMCA | 52.7% | 26.5% |
| Epoch Capabilities Index | 157.45 | 145.64 |
| ForecastBench | 61.1 | 61.8 |
| Kagi LLM Benchmark | — | 67.6% |
| NYT Connections (extended) | 93.6% | — |
| EnigmaEval | — | 9.2% |
| Surface Evolver Bench | 95% | — |
Math Kimi K3 leads
Kimi K3: 74.2 (#16), o4-mini: 40.8 (#89)
| Benchmark | Kimi K3 | o4-mini |
|---|---|---|
| FrontierMath (Tiers 1-3) | 72.2% | 36.1% |
| FrontierMath Tier 4 | 39% | 4.9% |
| OTIS Mock AIME 2024-2025 | 97.2% | 81.7% |
| LMArena Math | 1491 | 1389 |
| MathArena Final-Answer Competitions | 87.8% | — |
| ProofBench | 87% | — |
| Omni-MATH | — | 72% |
| MATH Level 5 | — | 97.8% |
| FrontierMath (Feb 2025 set) | — | 24.8% |
| FrontierMath Tier 4 (v1) | — | 6.3% |
Knowledge Kimi K3 leads
Kimi K3: 63.2 (#21), o4-mini: 43.6 (#91)
| Benchmark | Kimi K3 | o4-mini |
|---|---|---|
| GPQA Diamond | 93.1% | 79.6% |
| SimpleQA Verified | 50.6% | 19.6% |
| LMArena Expert | 1521 | 1343 |
| Humanity's Last Exam | — | 18.1% |
| MMLU-Pro | — | 82% |
| Confabulations | — | 15.8% |
| Vectara Hallucination Rate | — | 18.6% |
| GPQA (HELM) | — | 73.5% |
Multimodal o4-mini leads
Kimi K3: 37.8 (#70), o4-mini: 40.2 (#49)
| Benchmark | Kimi K3 | o4-mini |
|---|---|---|
| LMArena Vision | — | 1194 |
| GeoBench | — | 64% |
| VPCT | — | 57.5% |
| Blueprint-Bench 2 | 29.5% | — |
| Furniture Assembly | 34.2% | — |
Multilingual Kimi K3 leads
Kimi K3: 56.3 (#21), o4-mini: 47.0 (#154)
| Benchmark | Kimi K3 | o4-mini |
|---|---|---|
| LMArena Non-English | 1466 | 1337 |
| LMArena Chinese | 1529 | 1354 |
| LMArena French | 1491 | 1364 |
| LMArena German | 1488 | 1336 |
| LMArena Japanese | 1487 | 1308 |
| LMArena Korean | 1458 | 1312 |
| LMArena Russian | 1482 | 1334 |
| LMArena Spanish | 1472 | 1347 |
Instruction Following Kimi K3 leads
Kimi K3: 77.7 (#14), o4-mini: 75.2 (#68)
| Benchmark | Kimi K3 | o4-mini |
|---|---|---|
| LMArena Instruction Following | 1483 | 1321 |
| IFEval | — | 92.8% |
Long Context Too close to call
Kimi K3: 45.8 (#29), o4-mini: 45.5 (#33)
| Benchmark | Kimi K3 | o4-mini |
|---|---|---|
| LMArena Longer Query | 1494 | 1315 |
| Fiction.LiveBench | — | 77.8% |
Writing & Preference Kimi K3 leads
Kimi K3: 76.6 (#4), o4-mini: 54.0 (#152)
| Benchmark | Kimi K3 | o4-mini |
|---|---|---|
| LMArena Text | 1476 | 1353 |
| LMArena Creative Writing | 1454 | 1294 |
| LMArena Multi-Turn | 1488 | 1350 |
| Short-Story Creative Writing | — | 75% |
| EQ-Bench Creative Writing | 2082 | — |
| WildBench | — | 85.4% |
| EQ-Bench 4 | 1339 | — |
Frequently asked questions
Is Kimi K3 better than o4-mini?
Kimi K3 is the stronger model overall, scoring 59.5 to 41.6 on the Noometry Index. o4-mini costs 3.1× less per token, which makes it the better buy when Kimi K3's lead doesn't matter for your workload.
Which is cheaper, Kimi K3 or o4-mini?
o4-mini is cheaper. It lists at $1.10 per million input tokens and $4.40 per million output tokens; Kimi K3 lists at $3 and $15.
Is Kimi K3 or o4-mini better for coding?
Kimi K3 scores higher on coding benchmarks: 61.0 versus 40.9 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 o4-mini share?
34 benchmarks have published results for both models. Kimi K3 has 53 scored results on Noometry and o4-mini has 60.