# Kimi K3 vs MiniMax-M2.7

> Kimi K3 is the stronger model overall, scoring 59.5 to 37.7 on the Noometry Index. MiniMax-M2.7 costs 11× less per token, which makes it the better buy when Kimi K3's lead doesn't matter for your workload.

- Canonical page: https://noometry.com/compare/kimi-k3-vs-minimax-m2-7
- Last updated: 2026-10-10
- Shared benchmarks: 26

## Summary

- They share 26 benchmarks with published results for both. Kimi K3 scores higher in 9 categories and MiniMax-M2.7 in 0 categories; 9 gaps are clear of the uncertainty.
- The widest gap is in math, where Kimi K3 leads 74.2 to 25.9.
- The biggest single-benchmark swing is ProofBench: 87% for Kimi K3 and 3% for MiniMax-M2.7.
- MiniMax-M2.7 is cheaper at $0.30 / $1.20 per million input/output tokens, against $3 / $15 for Kimi K3.
- Kimi K3 accepts more context: 1.05M tokens versus 205K.

## Snapshot

| | Kimi K3 | MiniMax-M2.7 |
|---|---|---|
| Provider | Moonshot AI | MiniMax |
| Noometry Index | 59.5 | 37.7 |
| Rank | 15 | 196 |
| Context | 1.05M | 205K |
| Input $/M | $3 | $0.30 |
| Output $/M | $15 | $1.20 |
| Weights | Open | Open |

## Coding

- Kimi K3: 61.0 (#10)
- MiniMax-M2.7: 41.8 (#120)

| Benchmark | Kimi K3 | MiniMax-M2.7 |
|---|---|---|
| LMArena WebDev | 1654 | 1398 |
| SciCode | 59.5% | 47% |
| WeirdML | 82.6% | 37% |
| LMArena Coding | 1508 | 1454 |
| ALE-Bench | 1,524 | 599.25 |
| DeepSWE | 68.5% | — |
| FrontierCode | 44.2% | — |
| FrontierSWE | 25.9% | — |

## Agentic & Tool Use

- Kimi K3: 41.8 (#20)
- MiniMax-M2.7: 25.1 (#111)

| Benchmark | Kimi K3 | MiniMax-M2.7 |
|---|---|---|
| GBAEval | 48.3% | 0% |
| Terminal-Bench | — | 45.1% |
| APEX-Agents | 50.6% | — |
| τ²-bench Banking | 37.1% | — |
| PostTrainBench | 32% | — |
| ExploitBench | — | 13.3% |
| GDP.pdf | 19% | — |
| Vending-Bench 2 | 5,165 | — |

## Reasoning

- Kimi K3: 63.0 (#17)
- MiniMax-M2.7: 19.7 (#253)

| Benchmark | Kimi K3 | MiniMax-M2.7 |
|---|---|---|
| NYT Connections (extended) | 93.6% | 24.7% |
| CritPt | 23.4% | 0.6% |
| LMArena Hard Prompts | 1496 | 1422 |
| Epoch Capabilities Index | 157.45 | 145.85 |
| ARC-AGI-2 | 60.4% | — |
| SimpleBench | 60.7% | — |
| ARC-AGI-1 | 94.5% | — |
| Chess Puzzles | 39% | — |
| Thematic Generalization | — | 39.3% |
| Mystery Game Puzzles | 26% | — |
| DTBench | 91.2% | — |
| LMCA | 52.7% | — |
| Surface Evolver Bench | 95% | — |
| ForecastBench | 61.1 | — |

## Math

- Kimi K3: 74.2 (#16)
- MiniMax-M2.7: 25.9 (#263)

| Benchmark | Kimi K3 | MiniMax-M2.7 |
|---|---|---|
| ProofBench | 87% | 3% |
| LMArena Math | 1491 | 1420 |
| FrontierMath (Tiers 1-3) | 72.2% | — |
| FrontierMath Tier 4 | 39% | — |
| MathArena Final-Answer Competitions | 87.8% | — |
| OTIS Mock AIME 2024-2025 | 97.2% | — |

## Knowledge

- Kimi K3: 63.2 (#21)
- MiniMax-M2.7: 37.7 (#152)

| Benchmark | Kimi K3 | MiniMax-M2.7 |
|---|---|---|
| LMArena Expert | 1521 | 1444 |
| GPQA Diamond | 93.1% | — |
| SimpleQA Verified | 50.6% | — |
| Vectara Hallucination Rate | — | 12.9% |

## Multimodal

- Kimi K3: 37.8 (#70)
- MiniMax-M2.7: —

| Benchmark | Kimi K3 | MiniMax-M2.7 |
|---|---|---|
| Blueprint-Bench 2 | 29.5% | — |
| Furniture Assembly | 34.2% | — |

## Multilingual

- Kimi K3: 56.3 (#21)
- MiniMax-M2.7: 50.3 (#123)

| Benchmark | Kimi K3 | MiniMax-M2.7 |
|---|---|---|
| LMArena Non-English | 1466 | 1382 |
| LMArena Chinese | 1529 | 1441 |
| LMArena French | 1491 | 1421 |
| LMArena German | 1488 | 1398 |
| LMArena Japanese | 1487 | 1262 |
| LMArena Korean | 1458 | 1313 |
| LMArena Russian | 1482 | 1383 |
| LMArena Spanish | 1472 | 1403 |

## Instruction Following

- Kimi K3: 77.7 (#14)
- MiniMax-M2.7: 74.1 (#103)

| Benchmark | Kimi K3 | MiniMax-M2.7 |
|---|---|---|
| LMArena Instruction Following | 1483 | 1405 |

## Long Context

- Kimi K3: 45.8 (#29)
- MiniMax-M2.7: 43.3 (#99)

| Benchmark | Kimi K3 | MiniMax-M2.7 |
|---|---|---|
| LMArena Longer Query | 1494 | 1419 |

## Writing & Preference

- Kimi K3: 76.6 (#4)
- MiniMax-M2.7: 58.9 (#112)

| Benchmark | Kimi K3 | MiniMax-M2.7 |
|---|---|---|
| LMArena Text | 1476 | 1405 |
| LMArena Creative Writing | 1454 | 1354 |
| LMArena Multi-Turn | 1488 | 1412 |
| EQ-Bench Creative Writing | 2082 | — |
| EQ-Bench 4 | 1339 | — |

## FAQ

### Is Kimi K3 better than MiniMax-M2.7?

Kimi K3 is the stronger model overall, scoring 59.5 to 37.7 on the Noometry Index. MiniMax-M2.7 costs 11× 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 MiniMax-M2.7?

MiniMax-M2.7 is cheaper. It lists at $0.30 per million input tokens and $1.20 per million output tokens; Kimi K3 lists at $3 and $15.

### Is Kimi K3 or MiniMax-M2.7 better for coding?

Kimi K3 scores higher on coding benchmarks: 61.0 versus 41.8 in the Noometry coding category.

### Which has the bigger context window?

Kimi K3 does, with 1.05M tokens against 205K.

### How many benchmarks do Kimi K3 and MiniMax-M2.7 share?

26 benchmarks have published results for both models. Kimi K3 has 53 scored results on Noometry and MiniMax-M2.7 has 30.