Model comparison
Llama 3.1-70B vs Qwen2.5-Max
Qwen2.5-Max is the stronger model overall, scoring 40.7 to 29.6 on the Noometry Index.
Last verified . 18 shared benchmarks.
Meta
Alibaba (Qwen)Summary
- They share 18 benchmarks with published results for both. Llama 3.1-70B scores higher in 0 categories and Qwen2.5-Max in 8 categories; 8 gaps are clear of the uncertainty.
- The widest gap is in math, where Qwen2.5-Max leads 36.9 to 13.5.
- Llama 3.1-70B has downloadable open weights; the other is API-only.
Side by side
| Llama 3.1-70B | Qwen2.5-Max | |
|---|---|---|
| Provider | Meta | Alibaba (Qwen) |
| Noometry Index | 29.6 | 40.7 |
| Released | 2024-07-23 | 2025-01-25 |
| Weights | Open | Proprietary |
| Context window | 128K | — |
| Max output | 4K | — |
| Input $ / M tokens | $0.40 | — |
| Output $ / M tokens | $0.40 | — |
| Results tracked | 35 | 27 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding Qwen2.5-Max leads
Llama 3.1-70B: 30.3 (#296), Qwen2.5-Max: 41.8 (#117)
| Benchmark | Llama 3.1-70B | Qwen2.5-Max |
|---|---|---|
| LMArena Coding | 1260 | 1359 |
| WeirdML | 9% | — |
| BigCodeBench Instruct | 46.1% | — |
| LiveBench Coding | — | 64.4% |
| BigCodeBench Complete | 54.8% | — |
Agentic & Tool Use Not comparable
Llama 3.1-70B: 25.1 (#112), Qwen2.5-Max: —
| Benchmark | Llama 3.1-70B | Qwen2.5-Max |
|---|---|---|
| TheAgentCompany | 6.9% | — |
| BALROG | 27.9% | — |
Reasoning Qwen2.5-Max leads
Llama 3.1-70B: 21.6 (#220), Qwen2.5-Max: 25.6 (#147)
| Benchmark | Llama 3.1-70B | Qwen2.5-Max |
|---|---|---|
| LMArena Hard Prompts | 1241 | 1360 |
| Epoch Capabilities Index | 125.92 | 132.53 |
| LiveBench Reasoning | — | 51.4% |
| DTBench | 60% | — |
| LiveBench Data Analysis | — | 67.9% |
| LMCA | 14.8% | — |
| LiveBench | — | 62.3% |
Math Qwen2.5-Max leads
Llama 3.1-70B: 13.5 (#304), Qwen2.5-Max: 36.9 (#162)
| Benchmark | Llama 3.1-70B | Qwen2.5-Max |
|---|---|---|
| LMArena Math | 1252 | 1369 |
| OTIS Mock AIME 2024-2025 | 3.6% | — |
| Omni-MATH | 21% | — |
| LiveBench Math | — | 58.4% |
| MATH Level 5 | 36.7% | — |
Knowledge Qwen2.5-Max leads
Llama 3.1-70B: 24.2 (#269), Qwen2.5-Max: 35.3 (#186)
| Benchmark | Llama 3.1-70B | Qwen2.5-Max |
|---|---|---|
| LMArena Expert | 1209 | 1337 |
| GPQA Diamond | 44.2% | — |
| MMLU-Pro | 65.3% | — |
| Confabulations | — | 21.8% |
| GPQA (HELM) | 42.6% | — |
| MMLU | 80.1% | — |
Multilingual Qwen2.5-Max leads
Llama 3.1-70B: 38.8 (#225), Qwen2.5-Max: 48.1 (#146)
| Benchmark | Llama 3.1-70B | Qwen2.5-Max |
|---|---|---|
| LMArena Non-English | 1219 | 1352 |
| LMArena Chinese | 1215 | 1382 |
| LMArena French | 1261 | 1396 |
| LMArena German | 1222 | 1350 |
| LMArena Japanese | 1132 | 1300 |
| LMArena Korean | 1140 | 1304 |
| LMArena Russian | 1234 | 1353 |
| LMArena Spanish | 1253 | 1377 |
Instruction Following Qwen2.5-Max leads
Llama 3.1-70B: 65.3 (#223), Qwen2.5-Max: 71.3 (#152)
| Benchmark | Llama 3.1-70B | Qwen2.5-Max |
|---|---|---|
| LMArena Instruction Following | 1231 | 1335 |
| LiveBench Instruction Following | — | 75.3% |
| IFEval | 82.1% | — |
Long Context Qwen2.5-Max leads
Llama 3.1-70B: 37.6 (#214), Qwen2.5-Max: 41.4 (#142)
| Benchmark | Llama 3.1-70B | Qwen2.5-Max |
|---|---|---|
| LMArena Longer Query | 1241 | 1358 |
Writing & Preference Qwen2.5-Max leads
Llama 3.1-70B: 35.4 (#267), Qwen2.5-Max: 55.4 (#146)
| Benchmark | Llama 3.1-70B | Qwen2.5-Max |
|---|---|---|
| LMArena Text | 1261 | 1367 |
| LMArena Creative Writing | 1232 | 1339 |
| LMArena Multi-Turn | 1256 | 1364 |
| Short-Story Creative Writing | — | 72.9% |
| EQ-Bench Creative Writing | 784 | — |
| WildBench | 75.8% | — |
| LiveBench Language | — | 56.3% |
Frequently asked questions
Is Llama 3.1-70B better than Qwen2.5-Max?
Qwen2.5-Max is the stronger model overall, scoring 40.7 to 29.6 on the Noometry Index.
Is Llama 3.1-70B or Qwen2.5-Max better for coding?
Qwen2.5-Max scores higher on coding benchmarks: 41.8 versus 30.3 in the Noometry coding category.
How many benchmarks do Llama 3.1-70B and Qwen2.5-Max share?
18 benchmarks have published results for both models. Llama 3.1-70B has 35 scored results on Noometry and Qwen2.5-Max has 27.