Model comparison
Llama 3.1-405B vs Qwen2.5-Coder-32B
Qwen2.5-Coder-32B is the stronger model overall, scoring 33.4 to 30.7 on the Noometry Index.
Last verified . 17 shared benchmarks.
Meta
Alibaba (Qwen)Summary
- They share 17 benchmarks with published results for both. Llama 3.1-405B scores higher in 4 categories and Qwen2.5-Coder-32B in 4 categories; 7 gaps are clear of the uncertainty.
- The widest gap is in math, where Qwen2.5-Coder-32B leads 33.3 to 18.4.
Side by side
| Llama 3.1-405B | Qwen2.5-Coder-32B | |
|---|---|---|
| Provider | Meta | Alibaba (Qwen) |
| Noometry Index | 30.7 | 33.4 |
| Released | 2024-07-23 | 2024-09-18 |
| Weights | Open | Open |
| Context window | — | 33K |
| Max output | — | 29K |
| Input $ / M tokens | — | $0.66 |
| Output $ / M tokens | — | $1 |
| Results tracked | 42 | 31 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding Llama 3.1-405B leads
Llama 3.1-405B: 33.1 (#262), Qwen2.5-Coder-32B: 22.6 (#333)
| Benchmark | Llama 3.1-405B | Qwen2.5-Coder-32B |
|---|---|---|
| LMArena Coding | 1291 | 1276 |
| SWE-bench Verified (bash only) | — | 9% |
| Aider Polyglot | — | 16.4% |
| WeirdML | 21.4% | — |
| BigCodeBench Instruct | — | 49% |
| LiveBench Coding | — | 56.9% |
| BigCodeBench Complete | — | 58% |
| HumanEval+ | — | 87.2% |
| MBPP+ | — | 77% |
Agentic & Tool Use Not comparable
Llama 3.1-405B: 21.0 (#140), Qwen2.5-Coder-32B: —
| Benchmark | Llama 3.1-405B | Qwen2.5-Coder-32B |
|---|---|---|
| TheAgentCompany | 7.4% | — |
| Cybench | 7.5% | — |
Reasoning Qwen2.5-Coder-32B leads
Llama 3.1-405B: 16.8 (#300), Qwen2.5-Coder-32B: 21.2 (#225)
| Benchmark | Llama 3.1-405B | Qwen2.5-Coder-32B |
|---|---|---|
| LMArena Hard Prompts | 1269 | 1251 |
| Epoch Capabilities Index | 128.75 | 119.49 |
| HellaSwag | 89.2% | 83% |
| WinoGrande | 89.2% | 80.8% |
| SimpleBench | 23% | — |
| Kagi LLM Benchmark | 45% | — |
| LiveBench Reasoning | — | 42.1% |
| DTBench | 61.4% | — |
| LiveBench Data Analysis | — | 49.9% |
| BIG-Bench Hard | 82.9% | — |
| ForecastBench | 59.9 | — |
| LiveBench | — | 46.2% |
| PIQA | 85.9% | — |
Math Qwen2.5-Coder-32B leads
Llama 3.1-405B: 18.4 (#290), Qwen2.5-Coder-32B: 33.3 (#204)
| Benchmark | Llama 3.1-405B | Qwen2.5-Coder-32B |
|---|---|---|
| LMArena Math | 1281 | 1251 |
| OTIS Mock AIME 2024-2025 | 9.7% | — |
| Omni-MATH | 24.9% | — |
| LiveBench Math | — | 46.6% |
| MATH Level 5 | 49.8% | — |
| GSM8K | — | 93% |
Knowledge Qwen2.5-Coder-32B leads
Llama 3.1-405B: 30.4 (#227), Qwen2.5-Coder-32B: 33.4 (#203)
| Benchmark | Llama 3.1-405B | Qwen2.5-Coder-32B |
|---|---|---|
| LMArena Expert | 1243 | 1221 |
| ARC (AI2) Challenge | 95.3% | 70.5% |
| MMLU | 84.5% | 79.1% |
| GPQA Diamond | 50.9% | — |
| MMLU-Pro | 72.3% | — |
| Confabulations | 17.6% | — |
| GPQA (HELM) | 52.2% | — |
| TriviaQA | 82.7% | — |
Multilingual Llama 3.1-405B leads
Llama 3.1-405B: 40.7 (#214), Qwen2.5-Coder-32B: 37.8 (#235)
| Benchmark | Llama 3.1-405B | Qwen2.5-Coder-32B |
|---|---|---|
| LMArena Non-English | 1248 | 1205 |
| LMArena Chinese | 1242 | 1222 |
| LMArena Russian | 1265 | 1228 |
| LMArena French | 1279 | — |
| LMArena German | 1252 | — |
| LMArena Japanese | 1208 | — |
| LMArena Korean | 1184 | — |
| LMArena Spanish | 1260 | — |
Instruction Following Llama 3.1-405B leads
Llama 3.1-405B: 65.9 (#214), Qwen2.5-Coder-32B: 61.4 (#245)
| Benchmark | Llama 3.1-405B | Qwen2.5-Coder-32B |
|---|---|---|
| LMArena Instruction Following | 1259 | 1223 |
| LiveBench Instruction Following | — | 58.7% |
| IFEval | 81.1% | — |
Long Context Too close to call
Llama 3.1-405B: 38.4 (#197), Qwen2.5-Coder-32B: 38.0 (#208)
| Benchmark | Llama 3.1-405B | Qwen2.5-Coder-32B |
|---|---|---|
| LMArena Longer Query | 1266 | 1251 |
Writing & Preference Qwen2.5-Coder-32B leads
Llama 3.1-405B: 38.9 (#251), Qwen2.5-Coder-32B: 41.6 (#240)
| Benchmark | Llama 3.1-405B | Qwen2.5-Coder-32B |
|---|---|---|
| LMArena Text | 1284 | 1230 |
| LMArena Creative Writing | 1262 | 1174 |
| LMArena Multi-Turn | 1297 | 1222 |
| EQ-Bench Creative Writing | 870 | — |
| WildBench | 78.3% | — |
| LiveBench Language | — | 23.3% |
Frequently asked questions
Is Llama 3.1-405B better than Qwen2.5-Coder-32B?
Qwen2.5-Coder-32B is the stronger model overall, scoring 33.4 to 30.7 on the Noometry Index.
Is Llama 3.1-405B or Qwen2.5-Coder-32B better for coding?
Llama 3.1-405B scores higher on coding benchmarks: 33.1 versus 22.6 in the Noometry coding category.
How many benchmarks do Llama 3.1-405B and Qwen2.5-Coder-32B share?
17 benchmarks have published results for both models. Llama 3.1-405B has 42 scored results on Noometry and Qwen2.5-Coder-32B has 31.