Model comparison
Llama 3.1-405B vs Qwen2.5-Max
Qwen2.5-Max is the stronger model overall, scoring 40.1 to 30.4 on the Noometry Index.
Last verified . 19 shared benchmarks.
Meta
Alibaba (Qwen)Summary
- They share 19 benchmarks with published results for both. Llama 3.1-405B scores higher in 0 categories and Qwen2.5-Max in 8 categories; 8 gaps are clear of the uncertainty.
- The widest gap is in math, where Qwen2.5-Max leads 34.3 to 17.3.
- Llama 3.1-405B has downloadable open weights; the other is API-only.
Side by side
| Llama 3.1-405B | Qwen2.5-Max | |
|---|---|---|
| Provider | Meta | Alibaba (Qwen) |
| Noometry Index | 30.4 | 40.1 |
| Released | 2024-07-23 | 2025-01-25 |
| Weights | Open | Proprietary |
| Context window | — | — |
| Max output | — | — |
| Input $ / M tokens | — | — |
| Output $ / M tokens | — | — |
| Results tracked | 42 | 27 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding Qwen2.5-Max leads
Llama 3.1-405B: 32.7 (#257), Qwen2.5-Max: 41.2 (#118)
| Benchmark | Llama 3.1-405B | Qwen2.5-Max |
|---|---|---|
| LMArena Coding | 1291 | 1359 |
| WeirdML | 21.4% | — |
| LiveBench Coding | — | 64.4% |
Agentic & Tool Use Not comparable
Llama 3.1-405B: 20.9 (#131), Qwen2.5-Max: —
| Benchmark | Llama 3.1-405B | Qwen2.5-Max |
|---|---|---|
| TheAgentCompany | 7.4% | — |
| Cybench | 7.5% | — |
Reasoning Qwen2.5-Max leads
Llama 3.1-405B: 16.4 (#293), Qwen2.5-Max: 24.6 (#144)
| Benchmark | Llama 3.1-405B | Qwen2.5-Max |
|---|---|---|
| LMArena Hard Prompts | 1270 | 1360 |
| Epoch Capabilities Index | 128.76 | 132.54 |
| SimpleBench | 23% | — |
| Kagi LLM Benchmark | 45% | — |
| LiveBench Reasoning | — | 51.4% |
| DTBench | 61.4% | — |
| LiveBench Data Analysis | — | 67.9% |
| BIG-Bench Hard | 82.9% | — |
| ForecastBench | 59.9 | — |
| HellaSwag | 89.2% | — |
| LiveBench | — | 62.3% |
| PIQA | 85.9% | — |
| WinoGrande | 89.2% | — |
Math Qwen2.5-Max leads
Llama 3.1-405B: 17.3 (#283), Qwen2.5-Max: 34.3 (#162)
| Benchmark | Llama 3.1-405B | Qwen2.5-Max |
|---|---|---|
| LMArena Math | 1281 | 1369 |
| OTIS Mock AIME 2024-2025 | 9.7% | — |
| Omni-MATH | 24.9% | — |
| LiveBench Math | — | 58.4% |
| MATH Level 5 | 49.8% | — |
Knowledge Qwen2.5-Max leads
Llama 3.1-405B: 30.4 (#221), Qwen2.5-Max: 34.8 (#181)
| Benchmark | Llama 3.1-405B | Qwen2.5-Max |
|---|---|---|
| Confabulations | 17.6% | 21.8% |
| LMArena Expert | 1243 | 1337 |
| GPQA Diamond | 50.9% | — |
| MMLU-Pro | 72.3% | — |
| GPQA (HELM) | 52.2% | — |
| ARC (AI2) Challenge | 95.3% | — |
| MMLU | 84.5% | — |
| TriviaQA | 82.7% | — |
Multilingual Qwen2.5-Max leads
Llama 3.1-405B: 40.9 (#209), Qwen2.5-Max: 48.3 (#140)
| Benchmark | Llama 3.1-405B | Qwen2.5-Max |
|---|---|---|
| LMArena Non-English | 1248 | 1352 |
| LMArena Chinese | 1242 | 1382 |
| LMArena French | 1278 | 1396 |
| LMArena German | 1252 | 1350 |
| LMArena Japanese | 1208 | 1299 |
| LMArena Korean | 1184 | 1304 |
| LMArena Russian | 1265 | 1353 |
| LMArena Spanish | 1259 | 1377 |
Instruction Following Qwen2.5-Max leads
Llama 3.1-405B: 66.1 (#209), Qwen2.5-Max: 71.5 (#147)
| Benchmark | Llama 3.1-405B | Qwen2.5-Max |
|---|---|---|
| LMArena Instruction Following | 1259 | 1335 |
| LiveBench Instruction Following | — | 75.3% |
| IFEval | 81.1% | — |
Long Context Qwen2.5-Max leads
Llama 3.1-405B: 38.4 (#192), Qwen2.5-Max: 41.3 (#137)
| Benchmark | Llama 3.1-405B | Qwen2.5-Max |
|---|---|---|
| LMArena Longer Query | 1266 | 1358 |
Writing & Preference Qwen2.5-Max leads
Llama 3.1-405B: 39.1 (#246), Qwen2.5-Max: 55.6 (#141)
| Benchmark | Llama 3.1-405B | Qwen2.5-Max |
|---|---|---|
| LMArena Text | 1284 | 1367 |
| LMArena Creative Writing | 1262 | 1339 |
| LMArena Multi-Turn | 1297 | 1364 |
| Short-Story Creative Writing | — | 72.9% |
| EQ-Bench Creative Writing | 870 | — |
| WildBench | 78.3% | — |
| LiveBench Language | — | 56.3% |
Frequently asked questions
Is Llama 3.1-405B better than Qwen2.5-Max?
Qwen2.5-Max is the stronger model overall, scoring 40.1 to 30.4 on the Noometry Index.
Is Llama 3.1-405B or Qwen2.5-Max better for coding?
Qwen2.5-Max scores higher on coding benchmarks: 41.2 versus 32.7 in the Noometry coding category.
How many benchmarks do Llama 3.1-405B and Qwen2.5-Max share?
19 benchmarks have published results for both models. Llama 3.1-405B has 42 scored results on Noometry and Qwen2.5-Max has 27.