Model comparison
o3-pro vs Qwen3 32B
o3-pro is the stronger model overall, scoring 42.9 to 39.2 on the Noometry Index. Qwen3 32B costs 29× less per token, which makes it the better buy when o3-pro's lead doesn't matter for your workload.
Last verified . 7 shared benchmarks.
Alibaba (Qwen)Summary
- They share 7 benchmarks with published results for both. o3-pro scores higher in 4 categories and Qwen3 32B in 1 category; 5 gaps are clear of the uncertainty.
- The widest gap is in long context, where o3-pro leads 72.2 to 43.8.
- The biggest single-benchmark swing is Aider Polyglot: 84.9% for o3-pro and 40% for Qwen3 32B.
- Qwen3 32B is cheaper at $0.70 / $2.80 per million input/output tokens, against $20 / $80 for o3-pro.
- o3-pro accepts more context: 200K tokens versus 131K.
- Qwen3 32B has downloadable open weights; the other is API-only.
Side by side
| o3-pro | Qwen3 32B | |
|---|---|---|
| Provider | OpenAI | Alibaba (Qwen) |
| Noometry Index | 42.9 | 39.2 |
| Released | 2025-06-10 | 2025-04 |
| Weights | Proprietary | Open |
| Context window | 200K | 131K |
| Max output | 100K | 16K |
| Input $ / M tokens | $20 | $0.70 |
| Output $ / M tokens | $80 | $2.80 |
| Results tracked | 12 | 26 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding o3-pro leads
o3-pro: 55.5 (#24), Qwen3 32B: 37.7 (#190)
| Benchmark | o3-pro | Qwen3 32B |
|---|---|---|
| Aider Polyglot | 84.9% | 40% |
| SciCode | — | 35.4% |
| WeirdML | 58.2% | — |
| LMArena Coding | — | 1358 |
Agentic & Tool Use Not comparable
o3-pro: —, Qwen3 32B: 32.6 (#62)
| Benchmark | o3-pro | Qwen3 32B |
|---|---|---|
| Berkeley Function Calling Leaderboard | — | 48.7% |
Reasoning o3-pro leads
o3-pro: 23.8 (#171), Qwen3 32B: 20.2 (#241)
| Benchmark | o3-pro | Qwen3 32B |
|---|---|---|
| Kagi LLM Benchmark | 72.1% | 54.9% |
| DTBench | 86.9% | 67.5% |
| LMCA | 38.5% | 17.3% |
| Epoch Capabilities Index | 147.42 | 138.51 |
| ARC-AGI-2 | 4.9% | — |
| ARC-AGI-1 | 59.3% | — |
| CritPt | — | 0.3% |
| Chess Puzzles | — | 5% |
| LMArena Hard Prompts | — | 1334 |
Math Not comparable
o3-pro: —, Qwen3 32B: 39.7 (#99)
| Benchmark | o3-pro | Qwen3 32B |
|---|---|---|
| OTIS Mock AIME 2024-2025 | — | 66.9% |
| LMArena Math | — | 1399 |
Knowledge Qwen3 32B leads
o3-pro: 29.5 (#238), Qwen3 32B: 40.0 (#125)
| Benchmark | o3-pro | Qwen3 32B |
|---|---|---|
| Vectara Hallucination Rate | 23.3% | 5.9% |
| GPQA Diamond | — | 65.7% |
| Confabulations | 14.2% | — |
| LMArena Expert | — | 1362 |
Multilingual Not comparable
o3-pro: —, Qwen3 32B: 45.6 (#167)
| Benchmark | o3-pro | Qwen3 32B |
|---|---|---|
| LMArena Non-English | — | 1317 |
| LMArena Chinese | — | 1357 |
| LMArena German | — | 1341 |
| LMArena Russian | — | 1311 |
Instruction Following Not comparable
o3-pro: —, Qwen3 32B: 68.9 (#179)
| Benchmark | o3-pro | Qwen3 32B |
|---|---|---|
| LMArena Instruction Following | — | 1305 |
Long Context o3-pro leads
o3-pro: 72.2 (#1), Qwen3 32B: 43.8 (#87)
| Benchmark | o3-pro | Qwen3 32B |
|---|---|---|
| Fiction.LiveBench | 97.2% | 74.2% |
| LMArena Longer Query | — | 1327 |
Writing & Preference o3-pro leads
o3-pro: 57.1 (#133), Qwen3 32B: 52.9 (#163)
| Benchmark | o3-pro | Qwen3 32B |
|---|---|---|
| LMArena Text | — | 1340 |
| LMArena Creative Writing | — | 1297 |
| Short-Story Creative Writing | 84.4% | — |
| LMArena Multi-Turn | — | 1331 |
Frequently asked questions
Is o3-pro better than Qwen3 32B?
o3-pro is the stronger model overall, scoring 42.9 to 39.2 on the Noometry Index. Qwen3 32B costs 29× less per token, which makes it the better buy when o3-pro's lead doesn't matter for your workload.
Which is cheaper, o3-pro or Qwen3 32B?
Qwen3 32B is cheaper. It lists at $0.70 per million input tokens and $2.80 per million output tokens; o3-pro lists at $20 and $80.
Is o3-pro or Qwen3 32B better for coding?
o3-pro scores higher on coding benchmarks: 55.5 versus 37.7 in the Noometry coding category.
Which has the bigger context window?
o3-pro does, with 200K tokens against 131K.
How many benchmarks do o3-pro and Qwen3 32B share?
7 benchmarks have published results for both models. o3-pro has 12 scored results on Noometry and Qwen3 32B has 26.