# Llama 2-70B vs Qwen2.5-Max

> Qwen2.5-Max is the stronger model overall, scoring 40.1 to 24.1 on the Noometry Index.

- Canonical page: https://noometry.com/compare/llama-2-70b-vs-qwen2-5-max
- Last updated: 2026-10-10
- Shared benchmarks: 18

## Summary

- They share 18 benchmarks with published results for both. Llama 2-70B scores higher in 0 categories and Qwen2.5-Max in 8 categories; 8 gaps are clear of the uncertainty.
- The widest gap is in knowledge, where Qwen2.5-Max leads 34.8 to 7.4.
- Llama 2-70B has downloadable open weights; the other is API-only.

## Snapshot

| | Llama 2-70B | Qwen2.5-Max |
|---|---|---|
| Provider | Meta | Alibaba (Qwen) |
| Noometry Index | 24.1 | 40.1 |
| Rank | 343 | 142 |
| Context | — | — |
| Input $/M | — | — |
| Output $/M | — | — |
| Weights | Open | Proprietary |

## Coding

- Llama 2-70B: 30.8 (#281)
- Qwen2.5-Max: 41.2 (#118)

| Benchmark | Llama 2-70B | Qwen2.5-Max |
|---|---|---|
| LMArena Coding | 1079 | 1359 |
| LiveBench Coding | — | 64.4% |

## Reasoning

- Llama 2-70B: 13.7 (#324)
- Qwen2.5-Max: 24.6 (#144)

| Benchmark | Llama 2-70B | Qwen2.5-Max |
|---|---|---|
| LMArena Hard Prompts | 1073 | 1360 |
| Epoch Capabilities Index | 113.59 | 132.54 |
| LiveBench Reasoning | — | 51.4% |
| DTBench | 41.6% | — |
| LiveBench Data Analysis | — | 67.9% |
| BIG-Bench Hard | 64.9% | — |
| CommonsenseQA 2.0 | 50% | — |
| ForecastBench | 51.4 | — |
| HellaSwag | 85.3% | — |
| LAMBADA | 78.9% | — |
| LiveBench | — | 62.3% |
| PIQA | 82.8% | — |
| WinoGrande | 80.2% | — |

## Math

- Llama 2-70B: 7.5 (#320)
- Qwen2.5-Max: 34.3 (#162)

| Benchmark | Llama 2-70B | Qwen2.5-Max |
|---|---|---|
| LMArena Math | 1091 | 1369 |
| OTIS Mock AIME 2024-2025 | 0% | — |
| LiveBench Math | — | 58.4% |
| MATH Level 5 | 3.3% | — |
| GSM8K | 69.6% | — |

## Knowledge

- Llama 2-70B: 7.4 (#305)
- Qwen2.5-Max: 34.8 (#181)

| Benchmark | Llama 2-70B | Qwen2.5-Max |
|---|---|---|
| LMArena Expert | 1039 | 1337 |
| GPQA Diamond | 26.3% | — |
| Confabulations | — | 21.8% |
| ARC (AI2) Challenge | 78.3% | — |
| BoolQ | 88.6% | — |
| MMLU | 69.9% | — |
| OpenBookQA | 60.2% | — |
| TriviaQA | 87.6% | — |

## Multilingual

- Llama 2-70B: 27.9 (#269)
- Qwen2.5-Max: 48.3 (#140)

| Benchmark | Llama 2-70B | Qwen2.5-Max |
|---|---|---|
| LMArena Non-English | 1045 | 1352 |
| LMArena Chinese | 994 | 1382 |
| LMArena French | 1089 | 1396 |
| LMArena German | 1041 | 1350 |
| LMArena Japanese | 927 | 1299 |
| LMArena Korean | 964 | 1304 |
| LMArena Russian | 1083 | 1353 |
| LMArena Spanish | 1142 | 1377 |

## Instruction Following

- Llama 2-70B: 55.1 (#273)
- Qwen2.5-Max: 71.5 (#147)

| Benchmark | Llama 2-70B | Qwen2.5-Max |
|---|---|---|
| LMArena Instruction Following | 1071 | 1335 |
| LiveBench Instruction Following | — | 75.3% |

## Long Context

- Llama 2-70B: 32.2 (#265)
- Qwen2.5-Max: 41.3 (#137)

| Benchmark | Llama 2-70B | Qwen2.5-Max |
|---|---|---|
| LMArena Longer Query | 1062 | 1358 |

## Writing & Preference

- Llama 2-70B: 32.5 (#274)
- Qwen2.5-Max: 55.6 (#141)

| Benchmark | Llama 2-70B | Qwen2.5-Max |
|---|---|---|
| LMArena Text | 1115 | 1367 |
| LMArena Creative Writing | 1075 | 1339 |
| LMArena Multi-Turn | 1088 | 1364 |
| Short-Story Creative Writing | — | 72.9% |
| LiveBench Language | — | 56.3% |

## FAQ

### Is Llama 2-70B better than Qwen2.5-Max?

Qwen2.5-Max is the stronger model overall, scoring 40.1 to 24.1 on the Noometry Index.

### Is Llama 2-70B or Qwen2.5-Max better for coding?

Qwen2.5-Max scores higher on coding benchmarks: 41.2 versus 30.8 in the Noometry coding category.

### How many benchmarks do Llama 2-70B and Qwen2.5-Max share?

18 benchmarks have published results for both models. Llama 2-70B has 35 scored results on Noometry and Qwen2.5-Max has 27.