Model comparison
Qwen2.5-Coder-32B vs Qwen2.5-Max
Qwen2.5-Max is the stronger model overall, scoring 40.7 to 33.4 on the Noometry Index.
Last verified . 20 shared benchmarks.
Alibaba (Qwen)Summary
- They share 20 benchmarks with published results for both. Qwen2.5-Coder-32B scores higher in 0 categories and Qwen2.5-Max in 8 categories; 8 gaps are clear of the uncertainty.
- The widest gap is in coding, where Qwen2.5-Max leads 41.8 to 22.6.
- The biggest single-benchmark swing is LiveBench Language: 23.3% for Qwen2.5-Coder-32B and 56.3% for Qwen2.5-Max.
- Qwen2.5-Coder-32B has downloadable open weights; the other is API-only.
Side by side
| Qwen2.5-Coder-32B | Qwen2.5-Max | |
|---|---|---|
| Provider | Alibaba (Qwen) | Alibaba (Qwen) |
| Noometry Index | 33.4 | 40.7 |
| Released | 2024-09-18 | 2025-01-25 |
| Weights | Open | Proprietary |
| Context window | 33K | — |
| Max output | 29K | — |
| Input $ / M tokens | $0.66 | — |
| Output $ / M tokens | $1 | — |
| Results tracked | 31 | 27 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding Qwen2.5-Max leads
Qwen2.5-Coder-32B: 22.6 (#333), Qwen2.5-Max: 41.8 (#117)
| Benchmark | Qwen2.5-Coder-32B | Qwen2.5-Max |
|---|---|---|
| LiveBench Coding | 56.9% | 64.4% |
| LMArena Coding | 1276 | 1359 |
| SWE-bench Verified (bash only) | 9% | — |
| Aider Polyglot | 16.4% | — |
| BigCodeBench Instruct | 49% | — |
| BigCodeBench Complete | 58% | — |
| HumanEval+ | 87.2% | — |
| MBPP+ | 77% | — |
Reasoning Qwen2.5-Max leads
Qwen2.5-Coder-32B: 21.2 (#225), Qwen2.5-Max: 25.6 (#147)
| Benchmark | Qwen2.5-Coder-32B | Qwen2.5-Max |
|---|---|---|
| LiveBench Reasoning | 42.1% | 51.4% |
| LMArena Hard Prompts | 1251 | 1360 |
| LiveBench Data Analysis | 49.9% | 67.9% |
| Epoch Capabilities Index | 119.49 | 132.53 |
| LiveBench | 46.2% | 62.3% |
| HellaSwag | 83% | — |
| WinoGrande | 80.8% | — |
Math Qwen2.5-Max leads
Qwen2.5-Coder-32B: 33.3 (#204), Qwen2.5-Max: 36.9 (#162)
| Benchmark | Qwen2.5-Coder-32B | Qwen2.5-Max |
|---|---|---|
| LiveBench Math | 46.6% | 58.4% |
| LMArena Math | 1251 | 1369 |
| GSM8K | 93% | — |
Knowledge Qwen2.5-Max leads
Qwen2.5-Coder-32B: 33.4 (#203), Qwen2.5-Max: 35.3 (#186)
| Benchmark | Qwen2.5-Coder-32B | Qwen2.5-Max |
|---|---|---|
| LMArena Expert | 1221 | 1337 |
| Confabulations | — | 21.8% |
| ARC (AI2) Challenge | 70.5% | — |
| MMLU | 79.1% | — |
Multilingual Qwen2.5-Max leads
Qwen2.5-Coder-32B: 37.8 (#235), Qwen2.5-Max: 48.1 (#146)
| Benchmark | Qwen2.5-Coder-32B | Qwen2.5-Max |
|---|---|---|
| LMArena Non-English | 1205 | 1352 |
| LMArena Chinese | 1222 | 1382 |
| LMArena Russian | 1228 | 1353 |
| LMArena French | — | 1396 |
| LMArena German | — | 1350 |
| LMArena Japanese | — | 1300 |
| LMArena Korean | — | 1304 |
| LMArena Spanish | — | 1377 |
Instruction Following Qwen2.5-Max leads
Qwen2.5-Coder-32B: 61.4 (#245), Qwen2.5-Max: 71.3 (#152)
| Benchmark | Qwen2.5-Coder-32B | Qwen2.5-Max |
|---|---|---|
| LiveBench Instruction Following | 58.7% | 75.3% |
| LMArena Instruction Following | 1223 | 1335 |
Long Context Qwen2.5-Max leads
Qwen2.5-Coder-32B: 38.0 (#208), Qwen2.5-Max: 41.4 (#142)
| Benchmark | Qwen2.5-Coder-32B | Qwen2.5-Max |
|---|---|---|
| LMArena Longer Query | 1251 | 1358 |
Writing & Preference Qwen2.5-Max leads
Qwen2.5-Coder-32B: 41.6 (#240), Qwen2.5-Max: 55.4 (#146)
| Benchmark | Qwen2.5-Coder-32B | Qwen2.5-Max |
|---|---|---|
| LMArena Text | 1230 | 1367 |
| LMArena Creative Writing | 1174 | 1339 |
| LMArena Multi-Turn | 1222 | 1364 |
| LiveBench Language | 23.3% | 56.3% |
| Short-Story Creative Writing | — | 72.9% |
Frequently asked questions
Is Qwen2.5-Coder-32B better than Qwen2.5-Max?
Qwen2.5-Max is the stronger model overall, scoring 40.7 to 33.4 on the Noometry Index.
Is Qwen2.5-Coder-32B or Qwen2.5-Max better for coding?
Qwen2.5-Max scores higher on coding benchmarks: 41.8 versus 22.6 in the Noometry coding category.
How many benchmarks do Qwen2.5-Coder-32B and Qwen2.5-Max share?
20 benchmarks have published results for both models. Qwen2.5-Coder-32B has 31 scored results on Noometry and Qwen2.5-Max has 27.