# Llama 3-8B vs MiMo-V2.5-Pro

> MiMo-V2.5-Pro is the stronger model overall, scoring 44.8 to 25.3 on the Noometry Index. Llama 3-8B costs 3.9× less per token, which makes it the better buy when MiMo-V2.5-Pro's lead doesn't matter for your workload.

- Canonical page: https://noometry.com/compare/llama-3-8b-vs-mimo-v2-5-pro
- Last updated: 2026-10-10
- Shared benchmarks: 18

## Summary

- They share 18 benchmarks with published results for both. Llama 3-8B scores higher in 0 categories and MiMo-V2.5-Pro in 8 categories; 8 gaps are clear of the uncertainty.
- The widest gap is in knowledge, where MiMo-V2.5-Pro leads 41.5 to 7.8.
- The biggest single-benchmark swing is DTBench: 43.9% for Llama 3-8B and 84.5% for MiMo-V2.5-Pro.
- Llama 3-8B is cheaper at $0.14 / $0.14 per million input/output tokens, against $0.43 / $0.87 for MiMo-V2.5-Pro.
- MiMo-V2.5-Pro accepts more context: 1.05M tokens versus 8K.

## Snapshot

| | Llama 3-8B | MiMo-V2.5-Pro |
|---|---|---|
| Provider | Meta | Xiaomi |
| Noometry Index | 25.3 | 44.8 |
| Rank | 338 | 70 |
| Context | 8K | 1.05M |
| Input $/M | $0.14 | $0.43 |
| Output $/M | $0.14 | $0.87 |
| Weights | Open | Open |

## Coding

- Llama 3-8B: 30.5 (#285)
- MiMo-V2.5-Pro: 47.0 (#58)

| Benchmark | Llama 3-8B | MiMo-V2.5-Pro |
|---|---|---|
| LMArena Coding | 1152 | 1502 |
| LMArena WebDev | — | 1477 |
| SciCode | — | 50.2% |
| BigCodeBench Instruct | 31.9% | — |
| BigCodeBench Complete | 36.9% | — |
| ALE-Bench | — | 899.8 |
| HumanEval+ | 56.7% | — |
| MBPP+ | 54.8% | — |

## Reasoning

- Llama 3-8B: 13.8 (#321)
- MiMo-V2.5-Pro: 25.9 (#124)

| Benchmark | Llama 3-8B | MiMo-V2.5-Pro |
|---|---|---|
| LMArena Hard Prompts | 1133 | 1487 |
| DTBench | 43.9% | 84.5% |
| NYT Connections (extended) | — | 34.4% |
| CritPt | — | 4% |
| Chess Puzzles | 0% | — |
| LMCA | — | 29.5% |
| Adversarial NLI | 57.3% | — |
| Epoch Capabilities Index | 116.33 | — |
| ForecastBench | 58.6 | — |
| WinoGrande | 75.7% | — |

## Math

- Llama 3-8B: 8.3 (#317)
- MiMo-V2.5-Pro: 38.6 (#84)

| Benchmark | Llama 3-8B | MiMo-V2.5-Pro |
|---|---|---|
| LMArena Math | 1151 | 1480 |
| OTIS Mock AIME 2024-2025 | 1.9% | — |
| ProofBench | — | 22% |
| MATH Level 5 | 6.1% | — |

## Knowledge

- Llama 3-8B: 7.8 (#303)
- MiMo-V2.5-Pro: 41.5 (#100)

| Benchmark | Llama 3-8B | MiMo-V2.5-Pro |
|---|---|---|
| LMArena Expert | 1113 | 1501 |
| GPQA Diamond | 26.1% | — |
| ARC (AI2) Challenge | 82.8% | — |
| MMLU | 68.8% | — |
| OpenBookQA | 82.6% | — |
| TriviaQA | 67.7% | — |

## Multilingual

- Llama 3-8B: 31.0 (#256)
- MiMo-V2.5-Pro: 55.3 (#33)

| Benchmark | Llama 3-8B | MiMo-V2.5-Pro |
|---|---|---|
| LMArena Non-English | 1098 | 1449 |
| LMArena Chinese | 1076 | 1507 |
| LMArena French | 1158 | 1489 |
| LMArena German | 1104 | 1455 |
| LMArena Japanese | 967 | 1418 |
| LMArena Korean | 1003 | 1436 |
| LMArena Russian | 1109 | 1450 |
| LMArena Spanish | 1172 | 1470 |

## Instruction Following

- Llama 3-8B: 58.7 (#255)
- MiMo-V2.5-Pro: 77.6 (#21)

| Benchmark | Llama 3-8B | MiMo-V2.5-Pro |
|---|---|---|
| LMArena Instruction Following | 1127 | 1477 |

## Long Context

- Llama 3-8B: 34.2 (#246)
- MiMo-V2.5-Pro: 45.4 (#38)

| Benchmark | Llama 3-8B | MiMo-V2.5-Pro |
|---|---|---|
| LMArena Longer Query | 1128 | 1482 |

## Writing & Preference

- Llama 3-8B: 37.7 (#251)
- MiMo-V2.5-Pro: 65.5 (#47)

| Benchmark | Llama 3-8B | MiMo-V2.5-Pro |
|---|---|---|
| LMArena Text | 1166 | 1465 |
| LMArena Creative Writing | 1150 | 1440 |
| LMArena Multi-Turn | 1152 | 1477 |
| EQ-Bench Creative Writing | — | 1493 |
| EQ-Bench 4 | — | 1208 |

## FAQ

### Is Llama 3-8B better than MiMo-V2.5-Pro?

MiMo-V2.5-Pro is the stronger model overall, scoring 44.8 to 25.3 on the Noometry Index. Llama 3-8B costs 3.9× less per token, which makes it the better buy when MiMo-V2.5-Pro's lead doesn't matter for your workload.

### Which is cheaper, Llama 3-8B or MiMo-V2.5-Pro?

Llama 3-8B is cheaper. It lists at $0.14 per million input tokens and $0.14 per million output tokens; MiMo-V2.5-Pro lists at $0.43 and $0.87.

### Is Llama 3-8B or MiMo-V2.5-Pro better for coding?

MiMo-V2.5-Pro scores higher on coding benchmarks: 47.0 versus 30.5 in the Noometry coding category.

### Which has the bigger context window?

MiMo-V2.5-Pro does, with 1.05M tokens against 8K.

### How many benchmarks do Llama 3-8B and MiMo-V2.5-Pro share?

18 benchmarks have published results for both models. Llama 3-8B has 34 scored results on Noometry and MiMo-V2.5-Pro has 27.