Model comparison
Kimi K2 Thinking Turbo vs Kimi K3
Kimi K3 is the stronger model overall, scoring 59.5 to 45.8 on the Noometry Index.
Last verified . 21 shared benchmarks.
Summary
- They share 21 benchmarks with published results for both. Kimi K2 Thinking Turbo scores higher in 0 categories and Kimi K3 in 8 categories; 8 gaps are clear of the uncertainty.
- The widest gap is in reasoning, where Kimi K3 leads 63.0 to 32.5.
- The biggest single-benchmark swing is Chess Puzzles: 20% for Kimi K2 Thinking Turbo and 39% for Kimi K3.
Side by side
| Kimi K2 Thinking Turbo | Kimi K3 | |
|---|---|---|
| Provider | Moonshot AI | Moonshot AI |
| Noometry Index | 45.8 | 59.5 |
| Released | 2025-11-06 | 2026-07-16 |
| Weights | Open | Open |
| Context window | — | 1.05M |
| Max output | — | 1.05M |
| Input $ / M tokens | — | $3 |
| Output $ / M tokens | — | $15 |
| Results tracked | 21 | 53 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding Kimi K3 leads
Kimi K2 Thinking Turbo: 38.4 (#178), Kimi K3: 61.0 (#10)
| Benchmark | Kimi K2 Thinking Turbo | Kimi K3 |
|---|---|---|
| LMArena WebDev | 1322 | 1654 |
| LMArena Coding | 1454 | 1508 |
| DeepSWE | — | 68.5% |
| FrontierCode | — | 44.2% |
| FrontierSWE | — | 25.9% |
| SciCode | — | 59.5% |
| WeirdML | — | 82.6% |
| ALE-Bench | — | 1,524 |
Agentic & Tool Use Not comparable
Kimi K2 Thinking Turbo: —, Kimi K3: 41.8 (#20)
| Benchmark | Kimi K2 Thinking Turbo | Kimi K3 |
|---|---|---|
| APEX-Agents | — | 50.6% |
| τ²-bench Banking | — | 37.1% |
| PostTrainBench | — | 32% |
| GBAEval | — | 48.3% |
| GDP.pdf | — | 19% |
| Vending-Bench 2 | — | 5,165 |
Reasoning Kimi K3 leads
Kimi K2 Thinking Turbo: 32.5 (#75), Kimi K3: 63.0 (#17)
| Benchmark | Kimi K2 Thinking Turbo | Kimi K3 |
|---|---|---|
| Chess Puzzles | 20% | 39% |
| LMArena Hard Prompts | 1428 | 1496 |
| ARC-AGI-2 | — | 60.4% |
| SimpleBench | — | 60.7% |
| NYT Connections (extended) | — | 93.6% |
| ARC-AGI-1 | — | 94.5% |
| CritPt | — | 23.4% |
| Mystery Game Puzzles | — | 26% |
| DTBench | — | 91.2% |
| LMCA | — | 52.7% |
| Surface Evolver Bench | — | 95% |
| Epoch Capabilities Index | — | 157.45 |
| ForecastBench | — | 61.1 |
Math Kimi K3 leads
Kimi K2 Thinking Turbo: 47.4 (#68), Kimi K3: 74.2 (#16)
| Benchmark | Kimi K2 Thinking Turbo | Kimi K3 |
|---|---|---|
| OTIS Mock AIME 2024-2025 | 83.1% | 97.2% |
| LMArena Math | 1429 | 1491 |
| FrontierMath (Tiers 1-3) | — | 72.2% |
| FrontierMath Tier 4 | — | 39% |
| MathArena Final-Answer Competitions | — | 87.8% |
| ProofBench | — | 87% |
Knowledge Kimi K3 leads
Kimi K2 Thinking Turbo: 50.9 (#69), Kimi K3: 63.2 (#21)
| Benchmark | Kimi K2 Thinking Turbo | Kimi K3 |
|---|---|---|
| GPQA Diamond | 84.2% | 93.1% |
| LMArena Expert | 1439 | 1521 |
| SimpleQA Verified | — | 50.6% |
Multimodal Not comparable
Kimi K2 Thinking Turbo: —, Kimi K3: 37.8 (#70)
| Benchmark | Kimi K2 Thinking Turbo | Kimi K3 |
|---|---|---|
| Blueprint-Bench 2 | — | 29.5% |
| Furniture Assembly | — | 34.2% |
Multilingual Kimi K3 leads
Kimi K2 Thinking Turbo: 51.4 (#109), Kimi K3: 56.3 (#21)
| Benchmark | Kimi K2 Thinking Turbo | Kimi K3 |
|---|---|---|
| LMArena Non-English | 1398 | 1466 |
| LMArena Chinese | 1456 | 1529 |
| LMArena French | 1425 | 1491 |
| LMArena German | 1390 | 1488 |
| LMArena Japanese | 1357 | 1487 |
| LMArena Korean | 1330 | 1458 |
| LMArena Russian | 1391 | 1482 |
| LMArena Spanish | 1406 | 1472 |
Instruction Following Kimi K3 leads
Kimi K2 Thinking Turbo: 74.0 (#109), Kimi K3: 77.7 (#14)
| Benchmark | Kimi K2 Thinking Turbo | Kimi K3 |
|---|---|---|
| LMArena Instruction Following | 1403 | 1483 |
Long Context Kimi K3 leads
Kimi K2 Thinking Turbo: 43.2 (#102), Kimi K3: 45.8 (#29)
| Benchmark | Kimi K2 Thinking Turbo | Kimi K3 |
|---|---|---|
| LMArena Longer Query | 1415 | 1494 |
Writing & Preference Kimi K3 leads
Kimi K2 Thinking Turbo: 60.0 (#104), Kimi K3: 76.6 (#4)
| Benchmark | Kimi K2 Thinking Turbo | Kimi K3 |
|---|---|---|
| LMArena Text | 1415 | 1476 |
| LMArena Creative Writing | 1374 | 1454 |
| LMArena Multi-Turn | 1414 | 1488 |
| EQ-Bench Creative Writing | — | 2082 |
| EQ-Bench 4 | — | 1339 |
Frequently asked questions
Is Kimi K2 Thinking Turbo better than Kimi K3?
Kimi K3 is the stronger model overall, scoring 59.5 to 45.8 on the Noometry Index.
Is Kimi K2 Thinking Turbo or Kimi K3 better for coding?
Kimi K3 scores higher on coding benchmarks: 61.0 versus 38.4 in the Noometry coding category.
How many benchmarks do Kimi K2 Thinking Turbo and Kimi K3 share?
21 benchmarks have published results for both models. Kimi K2 Thinking Turbo has 21 scored results on Noometry and Kimi K3 has 53.