# Kimi K3 vs MiMo-V2.5

> Kimi K3 is the stronger model overall, scoring 59.5 to 43.4 on the Noometry Index. MiMo-V2.5 costs 34× 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-mimo-v2-5
- Last updated: 2026-10-10
- Shared benchmarks: 22

## Summary

- They share 22 benchmarks with published results for both. Kimi K3 scores higher in 8 categories and MiMo-V2.5 in 1 category; 9 gaps are clear of the uncertainty.
- The widest gap is in math, where Kimi K3 leads 74.2 to 36.8.
- The biggest single-benchmark swing is ProofBench: 87% for Kimi K3 and 16% for MiMo-V2.5.
- MiMo-V2.5 is cheaper at $0.14 / $0.28 per million input/output tokens, against $3 / $15 for Kimi K3.

## Snapshot

| | Kimi K3 | MiMo-V2.5 |
|---|---|---|
| Provider | Moonshot AI | Xiaomi |
| Noometry Index | 59.5 | 43.4 |
| Rank | 15 | 93 |
| Context | 1.05M | 1.05M |
| Input $/M | $3 | $0.14 |
| Output $/M | $15 | $0.28 |
| Weights | Open | Open |

## Coding

- Kimi K3: 61.0 (#10)
- MiMo-V2.5: 43.9 (#81)

| Benchmark | Kimi K3 | MiMo-V2.5 |
|---|---|---|
| LMArena WebDev | 1654 | 1438 |
| SciCode | 59.5% | 43.1% |
| LMArena Coding | 1508 | 1469 |
| ALE-Bench | 1,524 | 513.95 |
| DeepSWE | 68.5% | — |
| FrontierCode | 44.2% | — |
| FrontierSWE | 25.9% | — |
| WeirdML | 82.6% | — |

## Agentic & Tool Use

- Kimi K3: 41.8 (#20)
- MiMo-V2.5: —

| Benchmark | Kimi K3 | MiMo-V2.5 |
|---|---|---|
| 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: 63.0 (#17)
- MiMo-V2.5: 28.6 (#101)

| Benchmark | Kimi K3 | MiMo-V2.5 |
|---|---|---|
| CritPt | 23.4% | 3.7% |
| LMArena Hard Prompts | 1496 | 1450 |
| ARC-AGI-2 | 60.4% | — |
| SimpleBench | 60.7% | — |
| NYT Connections (extended) | 93.6% | — |
| ARC-AGI-1 | 94.5% | — |
| Chess Puzzles | 39% | — |
| 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: 74.2 (#16)
- MiMo-V2.5: 36.8 (#163)

| Benchmark | Kimi K3 | MiMo-V2.5 |
|---|---|---|
| ProofBench | 87% | 16% |
| LMArena Math | 1491 | 1436 |
| 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)
- MiMo-V2.5: 40.8 (#115)

| Benchmark | Kimi K3 | MiMo-V2.5 |
|---|---|---|
| LMArena Expert | 1521 | 1460 |
| GPQA Diamond | 93.1% | — |
| SimpleQA Verified | 50.6% | — |

## Multimodal

- Kimi K3: 37.8 (#70)
- MiMo-V2.5: 39.8 (#54)

| Benchmark | Kimi K3 | MiMo-V2.5 |
|---|---|---|
| LMArena Vision | — | 1247 |
| Blueprint-Bench 2 | 29.5% | — |
| Furniture Assembly | 34.2% | — |

## Multilingual

- Kimi K3: 56.3 (#21)
- MiMo-V2.5: 51.9 (#99)

| Benchmark | Kimi K3 | MiMo-V2.5 |
|---|---|---|
| LMArena Non-English | 1466 | 1404 |
| LMArena Chinese | 1529 | 1468 |
| LMArena French | 1491 | 1447 |
| LMArena German | 1488 | 1421 |
| LMArena Japanese | 1487 | 1306 |
| LMArena Korean | 1458 | 1363 |
| LMArena Russian | 1482 | 1395 |
| LMArena Spanish | 1472 | 1416 |

## Instruction Following

- Kimi K3: 77.7 (#14)
- MiMo-V2.5: 75.5 (#60)

| Benchmark | Kimi K3 | MiMo-V2.5 |
|---|---|---|
| LMArena Instruction Following | 1483 | 1434 |

## Long Context

- Kimi K3: 45.8 (#29)
- MiMo-V2.5: 44.2 (#73)

| Benchmark | Kimi K3 | MiMo-V2.5 |
|---|---|---|
| LMArena Longer Query | 1494 | 1445 |

## Writing & Preference

- Kimi K3: 76.6 (#4)
- MiMo-V2.5: 61.6 (#86)

| Benchmark | Kimi K3 | MiMo-V2.5 |
|---|---|---|
| LMArena Text | 1476 | 1428 |
| LMArena Creative Writing | 1454 | 1393 |
| LMArena Multi-Turn | 1488 | 1445 |
| EQ-Bench Creative Writing | 2082 | — |
| EQ-Bench 4 | 1339 | — |

## FAQ

### Is Kimi K3 better than MiMo-V2.5?

Kimi K3 is the stronger model overall, scoring 59.5 to 43.4 on the Noometry Index. MiMo-V2.5 costs 34× 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 MiMo-V2.5?

MiMo-V2.5 is cheaper. It lists at $0.14 per million input tokens and $0.28 per million output tokens; Kimi K3 lists at $3 and $15.

### Is Kimi K3 or MiMo-V2.5 better for coding?

Kimi K3 scores higher on coding benchmarks: 61.0 versus 43.9 in the Noometry coding category.

### Which has the bigger context window?

Both accept 1.05M tokens.

### How many benchmarks do Kimi K3 and MiMo-V2.5 share?

22 benchmarks have published results for both models. Kimi K3 has 53 scored results on Noometry and MiMo-V2.5 has 23.