Model comparison
o3 vs Qwen3 32B
o3 is the stronger model overall, scoring 47.5 to 39.2 on the Noometry Index. Qwen3 32B costs 2.9× less per token, which makes it the better buy when o3's lead doesn't matter for your workload.
Last verified . 24 shared benchmarks.
Alibaba (Qwen)Summary
- They share 24 benchmarks with published results for both. o3 scores higher in 9 categories and Qwen3 32B in 0 categories; 9 gaps are clear of the uncertainty.
- The widest gap is in knowledge, where o3 leads 54.6 to 40.0.
- The biggest single-benchmark swing is Aider Polyglot: 81.3% for o3 and 40% for Qwen3 32B.
- Qwen3 32B is cheaper at $0.70 / $2.80 per million input/output tokens, against $2 / $8 for o3.
- o3 accepts more context: 200K tokens versus 131K.
- Qwen3 32B has downloadable open weights; the other is API-only.
Side by side
| o3 | Qwen3 32B | |
|---|---|---|
| Provider | OpenAI | Alibaba (Qwen) |
| Noometry Index | 47.5 | 39.2 |
| Released | 2025-04-16 | 2025-04 |
| Weights | Proprietary | Open |
| Context window | 200K | 131K |
| Max output | 100K | 16K |
| Input $ / M tokens | $2 | $0.70 |
| Output $ / M tokens | $8 | $2.80 |
| Results tracked | 63 | 26 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding o3 leads
o3: 46.8 (#64), Qwen3 32B: 37.7 (#190)
| Benchmark | o3 | Qwen3 32B |
|---|---|---|
| Aider Polyglot | 81.3% | 40% |
| LMArena Coding | 1408 | 1358 |
| SWE-bench Verified | 62.3% | — |
| SWE-bench Verified (bash only) | 58.4% | — |
| SciCode | — | 35.4% |
| GSO | 8.8% | — |
| WeirdML | 52.4% | — |
| CadEval | 74% | — |
| ALE-Bench | 933.55 | — |
Agentic & Tool Use o3 leads
o3: 34.5 (#44), Qwen3 32B: 32.6 (#62)
| Benchmark | o3 | Qwen3 32B |
|---|---|---|
| Berkeley Function Calling Leaderboard | 63% | 48.7% |
| GDPval | 30.8% | — |
| DeepResearch Bench | 45.2% | — |
| OSWorld | 23% | — |
| LMArena Search | 1144 | — |
| METR Time Horizons | 65.4% | — |
Reasoning o3 leads
o3: 32.0 (#78), Qwen3 32B: 20.2 (#241)
| Benchmark | o3 | Qwen3 32B |
|---|---|---|
| Kagi LLM Benchmark | 67.6% | 54.9% |
| CritPt | 1.4% | 0.3% |
| Chess Puzzles | 38% | 5% |
| LMArena Hard Prompts | 1402 | 1334 |
| DTBench | 84.8% | 67.5% |
| LMCA | 39.7% | 17.3% |
| Epoch Capabilities Index | 146.86 | 138.51 |
| ARC-AGI-2 | 6.5% | — |
| SimpleBench | 53.1% | — |
| ARC-AGI-1 | 60.8% | — |
| EnigmaEval | 13.1% | — |
| Mystery Game Puzzles | 29% | — |
| ForecastBench | 62.5 | — |
Math o3 leads
o3: 50.2 (#58), Qwen3 32B: 39.7 (#99)
| Benchmark | o3 | Qwen3 32B |
|---|---|---|
| OTIS Mock AIME 2024-2025 | 84.4% | 66.9% |
| LMArena Math | 1426 | 1399 |
| FrontierMath (Tiers 1-3) | 33.3% | — |
| Omni-MATH | 71.4% | — |
| MATH Level 5 | 97.8% | — |
| FrontierMath (Feb 2025 set) | 18.7% | — |
| FrontierMath Tier 4 (v1) | 2.1% | — |
Knowledge o3 leads
o3: 54.6 (#52), Qwen3 32B: 40.0 (#125)
| Benchmark | o3 | Qwen3 32B |
|---|---|---|
| GPQA Diamond | 81.8% | 65.7% |
| LMArena Expert | 1402 | 1362 |
| Humanity's Last Exam | 20.3% | — |
| SimpleQA Verified | 49.4% | — |
| MMLU-Pro | 85.9% | — |
| Confabulations | 14.4% | — |
| Vectara Hallucination Rate | — | 5.9% |
| GPQA (HELM) | 75.3% | — |
Multimodal Not comparable
o3: 41.4 (#36), Qwen3 32B: —
| Benchmark | o3 | Qwen3 32B |
|---|---|---|
| LMArena Vision | 1214 | — |
| GeoBench | 74% | — |
| VPCT | 52% | — |
Multilingual o3 leads
o3: 51.7 (#105), Qwen3 32B: 45.6 (#167)
| Benchmark | o3 | Qwen3 32B |
|---|---|---|
| LMArena Non-English | 1401 | 1317 |
| LMArena Chinese | 1437 | 1357 |
| LMArena German | 1420 | 1341 |
| LMArena Russian | 1406 | 1311 |
| LMArena French | 1430 | — |
| LMArena Japanese | 1403 | — |
| LMArena Korean | 1370 | — |
| LMArena Spanish | 1395 | — |
Instruction Following o3 leads
o3: 72.8 (#127), Qwen3 32B: 68.9 (#179)
| Benchmark | o3 | Qwen3 32B |
|---|---|---|
| LMArena Instruction Following | 1368 | 1305 |
| IFEval | 86.9% | — |
Long Context o3 leads
o3: 53.3 (#6), Qwen3 32B: 43.8 (#87)
| Benchmark | o3 | Qwen3 32B |
|---|---|---|
| Fiction.LiveBench | 88.9% | 74.2% |
| LMArena Longer Query | 1372 | 1327 |
| CL-bench | 17.8% | — |
Writing & Preference o3 leads
o3: 63.5 (#64), Qwen3 32B: 52.9 (#163)
| Benchmark | o3 | Qwen3 32B |
|---|---|---|
| LMArena Text | 1410 | 1340 |
| LMArena Creative Writing | 1359 | 1297 |
| LMArena Multi-Turn | 1405 | 1331 |
| Short-Story Creative Writing | 83.9% | — |
| EQ-Bench Creative Writing | 1676 | — |
| WildBench | 86.1% | — |
Frequently asked questions
Is o3 better than Qwen3 32B?
o3 is the stronger model overall, scoring 47.5 to 39.2 on the Noometry Index. Qwen3 32B costs 2.9× less per token, which makes it the better buy when o3's lead doesn't matter for your workload.
Which is cheaper, o3 or Qwen3 32B?
Qwen3 32B is cheaper. It lists at $0.70 per million input tokens and $2.80 per million output tokens; o3 lists at $2 and $8.
Is o3 or Qwen3 32B better for coding?
o3 scores higher on coding benchmarks: 46.8 versus 37.7 in the Noometry coding category.
Which has the bigger context window?
o3 does, with 200K tokens against 131K.
How many benchmarks do o3 and Qwen3 32B share?
24 benchmarks have published results for both models. o3 has 63 scored results on Noometry and Qwen3 32B has 26.