Model comparison
Qwen2.5-Max vs Qwen3 32B
Qwen2.5-Max is the stronger model overall, scoring 40.7 to 39.2 on the Noometry Index.
Last verified . 14 shared benchmarks.
Alibaba (Qwen)Summary
- They share 14 benchmarks with published results for both. Qwen2.5-Max scores higher in 5 categories and Qwen3 32B in 3 categories; 8 gaps are clear of the uncertainty.
- The widest gap is in reasoning, where Qwen2.5-Max leads 25.6 to 20.2.
- Qwen3 32B has downloadable open weights; the other is API-only.
Side by side
| Qwen2.5-Max | Qwen3 32B | |
|---|---|---|
| Provider | Alibaba (Qwen) | Alibaba (Qwen) |
| Noometry Index | 40.7 | 39.2 |
| Released | 2025-01-25 | 2025-04 |
| Weights | Proprietary | Open |
| Context window | — | 131K |
| Max output | — | 16K |
| Input $ / M tokens | — | $0.70 |
| Output $ / M tokens | — | $2.80 |
| Results tracked | 27 | 26 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding Qwen2.5-Max leads
Qwen2.5-Max: 41.8 (#117), Qwen3 32B: 37.7 (#190)
| Benchmark | Qwen2.5-Max | Qwen3 32B |
|---|---|---|
| LMArena Coding | 1359 | 1358 |
| Aider Polyglot | — | 40% |
| SciCode | — | 35.4% |
| LiveBench Coding | 64.4% | — |
Agentic & Tool Use Not comparable
Qwen2.5-Max: —, Qwen3 32B: 32.6 (#62)
| Benchmark | Qwen2.5-Max | Qwen3 32B |
|---|---|---|
| Berkeley Function Calling Leaderboard | — | 48.7% |
Reasoning Qwen2.5-Max leads
Qwen2.5-Max: 25.6 (#147), Qwen3 32B: 20.2 (#241)
| Benchmark | Qwen2.5-Max | Qwen3 32B |
|---|---|---|
| LMArena Hard Prompts | 1360 | 1334 |
| Epoch Capabilities Index | 132.53 | 138.51 |
| Kagi LLM Benchmark | — | 54.9% |
| CritPt | — | 0.3% |
| Chess Puzzles | — | 5% |
| LiveBench Reasoning | 51.4% | — |
| DTBench | — | 67.5% |
| LiveBench Data Analysis | 67.9% | — |
| LMCA | — | 17.3% |
| LiveBench | 62.3% | — |
Math Qwen3 32B leads
Qwen2.5-Max: 36.9 (#162), Qwen3 32B: 39.7 (#99)
| Benchmark | Qwen2.5-Max | Qwen3 32B |
|---|---|---|
| LMArena Math | 1369 | 1399 |
| OTIS Mock AIME 2024-2025 | — | 66.9% |
| LiveBench Math | 58.4% | — |
Knowledge Qwen3 32B leads
Qwen2.5-Max: 35.3 (#186), Qwen3 32B: 40.0 (#125)
| Benchmark | Qwen2.5-Max | Qwen3 32B |
|---|---|---|
| LMArena Expert | 1337 | 1362 |
| GPQA Diamond | — | 65.7% |
| Confabulations | 21.8% | — |
| Vectara Hallucination Rate | — | 5.9% |
Multilingual Qwen2.5-Max leads
Qwen2.5-Max: 48.1 (#146), Qwen3 32B: 45.6 (#167)
| Benchmark | Qwen2.5-Max | Qwen3 32B |
|---|---|---|
| LMArena Non-English | 1352 | 1317 |
| LMArena Chinese | 1382 | 1357 |
| LMArena German | 1350 | 1341 |
| LMArena Russian | 1353 | 1311 |
| LMArena French | 1396 | — |
| LMArena Japanese | 1300 | — |
| LMArena Korean | 1304 | — |
| LMArena Spanish | 1377 | — |
Instruction Following Qwen2.5-Max leads
Qwen2.5-Max: 71.3 (#152), Qwen3 32B: 68.9 (#179)
| Benchmark | Qwen2.5-Max | Qwen3 32B |
|---|---|---|
| LMArena Instruction Following | 1335 | 1305 |
| LiveBench Instruction Following | 75.3% | — |
Long Context Qwen3 32B leads
Qwen2.5-Max: 41.4 (#142), Qwen3 32B: 43.8 (#87)
| Benchmark | Qwen2.5-Max | Qwen3 32B |
|---|---|---|
| LMArena Longer Query | 1358 | 1327 |
| Fiction.LiveBench | — | 74.2% |
Writing & Preference Qwen2.5-Max leads
Qwen2.5-Max: 55.4 (#146), Qwen3 32B: 52.9 (#163)
| Benchmark | Qwen2.5-Max | Qwen3 32B |
|---|---|---|
| LMArena Text | 1367 | 1340 |
| LMArena Creative Writing | 1339 | 1297 |
| LMArena Multi-Turn | 1364 | 1331 |
| Short-Story Creative Writing | 72.9% | — |
| LiveBench Language | 56.3% | — |
Frequently asked questions
Is Qwen2.5-Max better than Qwen3 32B?
Qwen2.5-Max is the stronger model overall, scoring 40.7 to 39.2 on the Noometry Index.
Is Qwen2.5-Max or Qwen3 32B better for coding?
Qwen2.5-Max scores higher on coding benchmarks: 41.8 versus 37.7 in the Noometry coding category.
How many benchmarks do Qwen2.5-Max and Qwen3 32B share?
14 benchmarks have published results for both models. Qwen2.5-Max has 27 scored results on Noometry and Qwen3 32B has 26.