Model comparison
o3-mini vs Qwen2.5-Coder-32B
o3-mini is the stronger model overall, scoring 36.7 to 33.4 on the Noometry Index. Qwen2.5-Coder-32B costs 2.6× less per token, which makes it the better buy when o3-mini's lead doesn't matter for your workload.
Last verified . 21 shared benchmarks.
Alibaba (Qwen)Summary
- They share 21 benchmarks with published results for both. o3-mini scores higher in 5 categories and Qwen2.5-Coder-32B in 3 categories; 8 gaps are clear of the uncertainty.
- The widest gap is in coding, where o3-mini leads 40.8 to 22.6.
- The biggest single-benchmark swing is LiveBench Reasoning: 89.6% for o3-mini and 42.1% for Qwen2.5-Coder-32B.
- Qwen2.5-Coder-32B is cheaper at $0.66 / $1 per million input/output tokens, against $1.10 / $4.40 for o3-mini.
- o3-mini accepts more context: 200K tokens versus 33K.
- Qwen2.5-Coder-32B has downloadable open weights; the other is API-only.
Side by side
| o3-mini | Qwen2.5-Coder-32B | |
|---|---|---|
| Provider | OpenAI | Alibaba (Qwen) |
| Noometry Index | 36.7 | 33.4 |
| Released | 2024-12-20 | 2024-09-18 |
| Weights | Proprietary | Open |
| Context window | 200K | 33K |
| Max output | 100K | 29K |
| Input $ / M tokens | $1.10 | $0.66 |
| Output $ / M tokens | $4.40 | $1 |
| Results tracked | 51 | 31 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding o3-mini leads
o3-mini: 40.8 (#132), Qwen2.5-Coder-32B: 22.6 (#333)
| Benchmark | o3-mini | Qwen2.5-Coder-32B |
|---|---|---|
| Aider Polyglot | 60.4% | 16.4% |
| LiveBench Coding | 82.7% | 56.9% |
| LMArena Coding | 1378 | 1276 |
| SWE-bench Verified (bash only) | — | 9% |
| SciCode | 39.8% | — |
| GSO | 1.3% | — |
| WeirdML | 43.7% | — |
| BigCodeBench Instruct | — | 49% |
| BigCodeBench Complete | — | 58% |
| CadEval | 54% | — |
| HumanEval+ | — | 87.2% |
| MBPP+ | — | 77% |
Agentic & Tool Use Not comparable
o3-mini: 29.6 (#84), Qwen2.5-Coder-32B: —
| Benchmark | o3-mini | Qwen2.5-Coder-32B |
|---|---|---|
| Cybench | 22.5% | — |
Reasoning Qwen2.5-Coder-32B leads
o3-mini: 16.3 (#305), Qwen2.5-Coder-32B: 21.2 (#225)
| Benchmark | o3-mini | Qwen2.5-Coder-32B |
|---|---|---|
| LiveBench Reasoning | 89.6% | 42.1% |
| LMArena Hard Prompts | 1366 | 1251 |
| LiveBench Data Analysis | 70.6% | 49.9% |
| Epoch Capabilities Index | 140.34 | 119.49 |
| LiveBench | 75.9% | 46.2% |
| ARC-AGI-2 | 3% | — |
| SimpleBench | 22.8% | — |
| ARC-AGI-1 | 34.5% | — |
| CritPt | 0.3% | — |
| Chess Puzzles | 17% | — |
| Mystery Game Puzzles | 7% | — |
| DTBench | 68.8% | — |
| LMCA | 19% | — |
| ForecastBench | 59.6 | — |
| HellaSwag | — | 83% |
| WinoGrande | — | 80.8% |
Math Qwen2.5-Coder-32B leads
o3-mini: 28.1 (#244), Qwen2.5-Coder-32B: 33.3 (#204)
| Benchmark | o3-mini | Qwen2.5-Coder-32B |
|---|---|---|
| LiveBench Math | 77.3% | 46.6% |
| LMArena Math | 1396 | 1251 |
| FrontierMath (Tiers 1-3) | 18.6% | — |
| FrontierMath Tier 4 | 0% | — |
| OTIS Mock AIME 2024-2025 | 76.9% | — |
| MATH Level 5 | 96.5% | — |
| FrontierMath (Feb 2025 set) | 12.4% | — |
| FrontierMath Tier 4 (v1) | 4.2% | — |
| GSM8K | — | 93% |
Knowledge o3-mini leads
o3-mini: 38.3 (#146), Qwen2.5-Coder-32B: 33.4 (#203)
| Benchmark | o3-mini | Qwen2.5-Coder-32B |
|---|---|---|
| LMArena Expert | 1364 | 1221 |
| GPQA Diamond | 77% | — |
| SimpleQA Verified | 15.3% | — |
| Confabulations | 17.9% | — |
| ARC (AI2) Challenge | — | 70.5% |
| MMLU | — | 79.1% |
Multilingual o3-mini leads
o3-mini: 45.7 (#164), Qwen2.5-Coder-32B: 37.8 (#235)
| Benchmark | o3-mini | Qwen2.5-Coder-32B |
|---|---|---|
| LMArena Non-English | 1319 | 1205 |
| LMArena Chinese | 1379 | 1222 |
| LMArena Russian | 1304 | 1228 |
| LMArena French | 1334 | — |
| LMArena German | 1303 | — |
| LMArena Japanese | 1286 | — |
| LMArena Korean | 1314 | — |
| LMArena Spanish | 1321 | — |
Instruction Following o3-mini leads
o3-mini: 75.1 (#72), Qwen2.5-Coder-32B: 61.4 (#245)
| Benchmark | o3-mini | Qwen2.5-Coder-32B |
|---|---|---|
| LiveBench Instruction Following | 84.4% | 58.7% |
| LMArena Instruction Following | 1337 | 1223 |
Long Context Qwen2.5-Coder-32B leads
o3-mini: 33.8 (#256), Qwen2.5-Coder-32B: 38.0 (#208)
| Benchmark | o3-mini | Qwen2.5-Coder-32B |
|---|---|---|
| LMArena Longer Query | 1343 | 1251 |
| Fiction.LiveBench | 50% | — |
Writing & Preference o3-mini leads
o3-mini: 50.3 (#182), Qwen2.5-Coder-32B: 41.6 (#240)
| Benchmark | o3-mini | Qwen2.5-Coder-32B |
|---|---|---|
| LMArena Text | 1337 | 1230 |
| LMArena Creative Writing | 1286 | 1174 |
| LMArena Multi-Turn | 1320 | 1222 |
| LiveBench Language | 50.7% | 23.3% |
| Short-Story Creative Writing | 61.7% | — |
Frequently asked questions
Is o3-mini better than Qwen2.5-Coder-32B?
o3-mini is the stronger model overall, scoring 36.7 to 33.4 on the Noometry Index. Qwen2.5-Coder-32B costs 2.6× less per token, which makes it the better buy when o3-mini's lead doesn't matter for your workload.
Which is cheaper, o3-mini or Qwen2.5-Coder-32B?
Qwen2.5-Coder-32B is cheaper. It lists at $0.66 per million input tokens and $1 per million output tokens; o3-mini lists at $1.10 and $4.40.
Is o3-mini or Qwen2.5-Coder-32B better for coding?
o3-mini scores higher on coding benchmarks: 40.8 versus 22.6 in the Noometry coding category.
Which has the bigger context window?
o3-mini does, with 200K tokens against 33K.
How many benchmarks do o3-mini and Qwen2.5-Coder-32B share?
21 benchmarks have published results for both models. o3-mini has 51 scored results on Noometry and Qwen2.5-Coder-32B has 31.