# Llama 3-70B vs Qwen2.5-Coder-32B

> Qwen2.5-Coder-32B is the stronger model overall, scoring 32.9 to 28.5 on the Noometry Index.

- Canonical page: https://noometry.com/compare/llama-3-70b-vs-qwen2-5-coder-32b
- Last updated: 2026-10-10
- Shared benchmarks: 19

## Summary

- They share 19 benchmarks with published results for both. Llama 3-70B scores higher in 3 categories and Qwen2.5-Coder-32B in 5 categories; 8 gaps are clear of the uncertainty.
- The widest gap is in math, where Qwen2.5-Coder-32B leads 30.8 to 12.0.
- The biggest single-benchmark swing is BigCodeBench Instruct: 43.6% for Llama 3-70B and 49% for Qwen2.5-Coder-32B.

## Snapshot

| | Llama 3-70B | Qwen2.5-Coder-32B |
|---|---|---|
| Provider | Meta | Alibaba (Qwen) |
| Noometry Index | 28.5 | 32.9 |
| Rank | 317 | 239 |
| Context | — | 33K |
| Input $/M | — | $0.66 |
| Output $/M | — | $1 |
| Weights | Open | Open |

## Coding

- Llama 3-70B: 35.3 (#215)
- Qwen2.5-Coder-32B: 22.4 (#328)

| Benchmark | Llama 3-70B | Qwen2.5-Coder-32B |
|---|---|---|
| BigCodeBench Instruct | 43.6% | 49% |
| LMArena Coding | 1206 | 1276 |
| BigCodeBench Complete | 54.5% | 58% |
| HumanEval+ | 72% | 87.2% |
| MBPP+ | 69% | 77% |
| SWE-bench Verified (bash only) | — | 9% |
| Aider Polyglot | — | 16.4% |
| LiveBench Coding | — | 56.9% |

## Agentic & Tool Use

- Llama 3-70B: 20.9 (#132)
- Qwen2.5-Coder-32B: —

| Benchmark | Llama 3-70B | Qwen2.5-Coder-32B |
|---|---|---|
| Cybench | 5% | — |

## Reasoning

- Llama 3-70B: 17.4 (#284)
- Qwen2.5-Coder-32B: 20.4 (#220)

| Benchmark | Llama 3-70B | Qwen2.5-Coder-32B |
|---|---|---|
| LMArena Hard Prompts | 1195 | 1251 |
| Epoch Capabilities Index | 122.91 | 119.41 |
| WinoGrande | 83.5% | 80.8% |
| Kagi LLM Benchmark | 35.1% | — |
| LiveBench Reasoning | — | 42.1% |
| DTBench | 54.2% | — |
| LiveBench Data Analysis | — | 49.9% |
| ForecastBench | 57.1 | — |
| HellaSwag | — | 83% |
| LiveBench | — | 46.2% |

## Math

- Llama 3-70B: 12.0 (#299)
- Qwen2.5-Coder-32B: 30.8 (#203)

| Benchmark | Llama 3-70B | Qwen2.5-Coder-32B |
|---|---|---|
| LMArena Math | 1218 | 1251 |
| OTIS Mock AIME 2024-2025 | 4.3% | — |
| LiveBench Math | — | 46.6% |
| MATH Level 5 | 22.6% | — |
| GSM8K | — | 93% |

## Knowledge

- Llama 3-70B: 20.8 (#272)
- Qwen2.5-Coder-32B: 32.9 (#200)

| Benchmark | Llama 3-70B | Qwen2.5-Coder-32B |
|---|---|---|
| LMArena Expert | 1149 | 1221 |
| MMLU | 79.3% | 79.1% |
| GPQA Diamond | 40.6% | — |
| ARC (AI2) Challenge | — | 70.5% |

## Multilingual

- Llama 3-70B: 33.8 (#246)
- Qwen2.5-Coder-32B: 38.0 (#230)

| Benchmark | Llama 3-70B | Qwen2.5-Coder-32B |
|---|---|---|
| LMArena Non-English | 1142 | 1205 |
| LMArena Chinese | 1114 | 1222 |
| LMArena Russian | 1159 | 1228 |
| LMArena French | 1231 | — |
| LMArena German | 1169 | — |
| LMArena Japanese | 1017 | — |
| LMArena Korean | 1017 | — |
| LMArena Spanish | 1241 | — |

## Instruction Following

- Llama 3-70B: 62.8 (#233)
- Qwen2.5-Coder-32B: 61.6 (#240)

| Benchmark | Llama 3-70B | Qwen2.5-Coder-32B |
|---|---|---|
| LMArena Instruction Following | 1194 | 1223 |
| LiveBench Instruction Following | — | 58.7% |

## Long Context

- Llama 3-70B: 35.6 (#235)
- Qwen2.5-Coder-32B: 37.9 (#204)

| Benchmark | Llama 3-70B | Qwen2.5-Coder-32B |
|---|---|---|
| LMArena Longer Query | 1174 | 1251 |

## Writing & Preference

- Llama 3-70B: 43.0 (#225)
- Qwen2.5-Coder-32B: 41.8 (#235)

| Benchmark | Llama 3-70B | Qwen2.5-Coder-32B |
|---|---|---|
| LMArena Text | 1221 | 1230 |
| LMArena Creative Writing | 1210 | 1174 |
| LMArena Multi-Turn | 1223 | 1222 |
| LiveBench Language | — | 23.3% |

## FAQ

### Is Llama 3-70B better than Qwen2.5-Coder-32B?

Qwen2.5-Coder-32B is the stronger model overall, scoring 32.9 to 28.5 on the Noometry Index.

### Is Llama 3-70B or Qwen2.5-Coder-32B better for coding?

Llama 3-70B scores higher on coding benchmarks: 35.3 versus 22.4 in the Noometry coding category.

### How many benchmarks do Llama 3-70B and Qwen2.5-Coder-32B share?

19 benchmarks have published results for both models. Llama 3-70B has 31 scored results on Noometry and Qwen2.5-Coder-32B has 31.