# Kimi K2 Thinking Turbo vs MiniMax-M3

> Kimi K2 Thinking Turbo is the stronger model overall, scoring 45.8 to 43.8 on the Noometry Index.

- Canonical page: https://noometry.com/compare/kimi-k2-thinking-turbo-vs-minimax-m3
- Last updated: 2026-10-12
- Shared benchmarks: 21

## Summary

- They share 21 benchmarks with published results for both. Kimi K2 Thinking Turbo scores higher in 2 categories and MiniMax-M3 in 6 categories; 7 gaps are clear of the uncertainty.
- The widest gap is in knowledge, where MiniMax-M3 leads 58.4 to 50.9.
- The biggest single-benchmark swing is OTIS Mock AIME 2024-2025: 83.1% for Kimi K2 Thinking Turbo and 71.1% for MiniMax-M3.

## Snapshot

| | Kimi K2 Thinking Turbo | MiniMax-M3 |
|---|---|---|
| Provider | Moonshot AI | MiniMax |
| Noometry Index | 45.8 | 43.8 |
| Rank | 70 | 85 |
| Context | — | 1M |
| Input $/M | — | $0.30 |
| Output $/M | — | $1.20 |
| Weights | Open | Open |

## Coding

- Kimi K2 Thinking Turbo: 38.4 (#178)
- MiniMax-M3: 41.8 (#118)

| Benchmark | Kimi K2 Thinking Turbo | MiniMax-M3 |
|---|---|---|
| LMArena WebDev | 1322 | 1482 |
| LMArena Coding | 1454 | 1469 |
| FrontierCode | — | 14.7% |
| SciCode | — | 47.1% |
| ALE-Bench | — | 640.02 |

## Agentic & Tool Use

- Kimi K2 Thinking Turbo: —
- MiniMax-M3: 22.6 (#130)

| Benchmark | Kimi K2 Thinking Turbo | MiniMax-M3 |
|---|---|---|
| APEX-Agents | — | 37.7% |
| OSWorld 2.0 | — | 4.6% |
| GBAEval | — | 0.9% |
| Vending-Bench 2 | — | 2,158 |

## Reasoning

- Kimi K2 Thinking Turbo: 32.5 (#75)
- MiniMax-M3: 30.1 (#87)

| Benchmark | Kimi K2 Thinking Turbo | MiniMax-M3 |
|---|---|---|
| Chess Puzzles | 20% | 14% |
| LMArena Hard Prompts | 1428 | 1447 |
| SimpleBench | — | 45.8% |
| NYT Connections (extended) | — | 65.1% |
| CritPt | — | 3.7% |
| Mystery Game Puzzles | — | 8% |
| DTBench | — | 78.9% |
| LMCA | — | 33.7% |
| Surface Evolver Bench | — | 55% |
| Epoch Capabilities Index | — | 146.95 |
| ForecastBench | — | 61.4 |

## Math

- Kimi K2 Thinking Turbo: 47.4 (#68)
- MiniMax-M3: 40.0 (#95)

| Benchmark | Kimi K2 Thinking Turbo | MiniMax-M3 |
|---|---|---|
| OTIS Mock AIME 2024-2025 | 83.1% | 71.1% |
| LMArena Math | 1429 | 1429 |
| ProofBench | — | 18% |

## Knowledge

- Kimi K2 Thinking Turbo: 50.9 (#69)
- MiniMax-M3: 58.4 (#35)

| Benchmark | Kimi K2 Thinking Turbo | MiniMax-M3 |
|---|---|---|
| GPQA Diamond | 84.2% | 90.9% |
| LMArena Expert | 1439 | 1461 |

## Multimodal

- Kimi K2 Thinking Turbo: —
- MiniMax-M3: 40.2 (#51)

| Benchmark | Kimi K2 Thinking Turbo | MiniMax-M3 |
|---|---|---|
| LMArena Vision | — | 1253 |
| LMArena Document | — | 1435 |

## Multilingual

- Kimi K2 Thinking Turbo: 51.4 (#109)
- MiniMax-M3: 53.0 (#75)

| Benchmark | Kimi K2 Thinking Turbo | MiniMax-M3 |
|---|---|---|
| LMArena Non-English | 1398 | 1420 |
| LMArena Chinese | 1456 | 1463 |
| LMArena French | 1425 | 1447 |
| LMArena German | 1390 | 1426 |
| LMArena Japanese | 1357 | 1381 |
| LMArena Korean | 1330 | 1372 |
| LMArena Russian | 1391 | 1428 |
| LMArena Spanish | 1406 | 1432 |

## Instruction Following

- Kimi K2 Thinking Turbo: 74.0 (#109)
- MiniMax-M3: 75.5 (#62)

| Benchmark | Kimi K2 Thinking Turbo | MiniMax-M3 |
|---|---|---|
| LMArena Instruction Following | 1403 | 1433 |

## Long Context

- Kimi K2 Thinking Turbo: 43.2 (#102)
- MiniMax-M3: 44.2 (#72)

| Benchmark | Kimi K2 Thinking Turbo | MiniMax-M3 |
|---|---|---|
| LMArena Longer Query | 1415 | 1445 |

## Writing & Preference

- Kimi K2 Thinking Turbo: 60.0 (#104)
- MiniMax-M3: 62.1 (#83)

| Benchmark | Kimi K2 Thinking Turbo | MiniMax-M3 |
|---|---|---|
| LMArena Text | 1415 | 1433 |
| LMArena Creative Writing | 1374 | 1404 |
| LMArena Multi-Turn | 1414 | 1442 |
| EQ-Bench 4 | — | 1150 |

## FAQ

### Is Kimi K2 Thinking Turbo better than MiniMax-M3?

Kimi K2 Thinking Turbo is the stronger model overall, scoring 45.8 to 43.8 on the Noometry Index.

### Is Kimi K2 Thinking Turbo or MiniMax-M3 better for coding?

MiniMax-M3 scores higher on coding benchmarks: 41.8 versus 38.4 in the Noometry coding category.

### How many benchmarks do Kimi K2 Thinking Turbo and MiniMax-M3 share?

21 benchmarks have published results for both models. Kimi K2 Thinking Turbo has 21 scored results on Noometry and MiniMax-M3 has 41.