Model comparison
Llama 2-70B vs Qwen2.5-Coder-32B
Qwen2.5-Coder-32B is the stronger model overall, scoring 33.4 to 24.4 on the Noometry Index.
Last verified . 18 shared benchmarks.
Meta
Alibaba (Qwen)Summary
- They share 18 benchmarks with published results for both. Llama 2-70B scores higher in 1 category and Qwen2.5-Coder-32B in 7 categories; 8 gaps are clear of the uncertainty.
- The widest gap is in knowledge, where Qwen2.5-Coder-32B leads 33.4 to 7.4.
Side by side
| Llama 2-70B | Qwen2.5-Coder-32B | |
|---|---|---|
| Provider | Meta | Alibaba (Qwen) |
| Noometry Index | 24.4 | 33.4 |
| Released | 2023-07-18 | 2024-09-18 |
| Weights | Open | Open |
| Context window | — | 33K |
| Max output | — | 29K |
| Input $ / M tokens | — | $0.66 |
| Output $ / M tokens | — | $1 |
| Results tracked | 35 | 31 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding Llama 2-70B leads
Llama 2-70B: 31.4 (#286), Qwen2.5-Coder-32B: 22.6 (#333)
| Benchmark | Llama 2-70B | Qwen2.5-Coder-32B |
|---|---|---|
| LMArena Coding | 1079 | 1276 |
| SWE-bench Verified (bash only) | — | 9% |
| Aider Polyglot | — | 16.4% |
| BigCodeBench Instruct | — | 49% |
| LiveBench Coding | — | 56.9% |
| BigCodeBench Complete | — | 58% |
| HumanEval+ | — | 87.2% |
| MBPP+ | — | 77% |
Reasoning Qwen2.5-Coder-32B leads
Llama 2-70B: 14.4 (#325), Qwen2.5-Coder-32B: 21.2 (#225)
| Benchmark | Llama 2-70B | Qwen2.5-Coder-32B |
|---|---|---|
| LMArena Hard Prompts | 1073 | 1251 |
| Epoch Capabilities Index | 113.79 | 119.49 |
| HellaSwag | 85.3% | 83% |
| WinoGrande | 80.2% | 80.8% |
| LiveBench Reasoning | — | 42.1% |
| DTBench | 41.6% | — |
| LiveBench Data Analysis | — | 49.9% |
| BIG-Bench Hard | 64.9% | — |
| CommonsenseQA 2.0 | 50% | — |
| ForecastBench | 51.4 | — |
| LAMBADA | 78.9% | — |
| LiveBench | — | 46.2% |
| PIQA | 82.8% | — |
Math Qwen2.5-Coder-32B leads
Llama 2-70B: 8.1 (#326), Qwen2.5-Coder-32B: 33.3 (#204)
| Benchmark | Llama 2-70B | Qwen2.5-Coder-32B |
|---|---|---|
| LMArena Math | 1091 | 1251 |
| GSM8K | 69.6% | 93% |
| OTIS Mock AIME 2024-2025 | 0% | — |
| LiveBench Math | — | 46.6% |
| MATH Level 5 | 3.3% | — |
Knowledge Qwen2.5-Coder-32B leads
Llama 2-70B: 7.4 (#310), Qwen2.5-Coder-32B: 33.4 (#203)
| Benchmark | Llama 2-70B | Qwen2.5-Coder-32B |
|---|---|---|
| LMArena Expert | 1039 | 1221 |
| ARC (AI2) Challenge | 78.3% | 70.5% |
| MMLU | 69.9% | 79.1% |
| GPQA Diamond | 26.3% | — |
| BoolQ | 88.6% | — |
| OpenBookQA | 60.2% | — |
| TriviaQA | 87.6% | — |
Multilingual Qwen2.5-Coder-32B leads
Llama 2-70B: 27.7 (#274), Qwen2.5-Coder-32B: 37.8 (#235)
| Benchmark | Llama 2-70B | Qwen2.5-Coder-32B |
|---|---|---|
| LMArena Non-English | 1045 | 1205 |
| LMArena Chinese | 995 | 1222 |
| LMArena Russian | 1083 | 1228 |
| LMArena French | 1090 | — |
| LMArena German | 1041 | — |
| LMArena Japanese | 927 | — |
| LMArena Korean | 964 | — |
| LMArena Spanish | 1143 | — |
Instruction Following Qwen2.5-Coder-32B leads
Llama 2-70B: 54.9 (#278), Qwen2.5-Coder-32B: 61.4 (#245)
| Benchmark | Llama 2-70B | Qwen2.5-Coder-32B |
|---|---|---|
| LMArena Instruction Following | 1071 | 1223 |
| LiveBench Instruction Following | — | 58.7% |
Long Context Qwen2.5-Coder-32B leads
Llama 2-70B: 32.3 (#270), Qwen2.5-Coder-32B: 38.0 (#208)
| Benchmark | Llama 2-70B | Qwen2.5-Coder-32B |
|---|---|---|
| LMArena Longer Query | 1062 | 1251 |
Writing & Preference Qwen2.5-Coder-32B leads
Llama 2-70B: 32.3 (#279), Qwen2.5-Coder-32B: 41.6 (#240)
| Benchmark | Llama 2-70B | Qwen2.5-Coder-32B |
|---|---|---|
| LMArena Text | 1115 | 1230 |
| LMArena Creative Writing | 1075 | 1174 |
| LMArena Multi-Turn | 1088 | 1222 |
| LiveBench Language | — | 23.3% |
Frequently asked questions
Is Llama 2-70B better than Qwen2.5-Coder-32B?
Qwen2.5-Coder-32B is the stronger model overall, scoring 33.4 to 24.4 on the Noometry Index.
Is Llama 2-70B or Qwen2.5-Coder-32B better for coding?
Llama 2-70B scores higher on coding benchmarks: 31.4 versus 22.6 in the Noometry coding category.
How many benchmarks do Llama 2-70B and Qwen2.5-Coder-32B share?
18 benchmarks have published results for both models. Llama 2-70B has 35 scored results on Noometry and Qwen2.5-Coder-32B has 31.