Model comparison
GPT-5 Mini vs QwQ-32B
GPT-5 Mini is the stronger model overall, scoring 41.8 to 39.8 on the Noometry Index.
Last verified . 26 shared benchmarks.
Alibaba (Qwen)Summary
- They share 26 benchmarks with published results for both. GPT-5 Mini scores higher in 7 categories and QwQ-32B in 1 category; 7 gaps are clear of the uncertainty.
- The widest gap is in math, where GPT-5 Mini leads 46.7 to 38.0.
- The biggest single-benchmark swing is OTIS Mock AIME 2024-2025: 86.7% for GPT-5 Mini and 59.2% for QwQ-32B.
- QwQ-32B has downloadable open weights; the other is API-only.
Side by side
| GPT-5 Mini | QwQ-32B | |
|---|---|---|
| Provider | OpenAI | Alibaba (Qwen) |
| Noometry Index | 41.8 | 39.8 |
| Released | 2025-08-07 | 2024-11-28 |
| Weights | Proprietary | Open |
| Context window | 400K | — |
| Max output | 128K | — |
| Input $ / M tokens | $0.25 | — |
| Output $ / M tokens | $2 | — |
| Results tracked | 60 | 36 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding GPT-5 Mini leads
GPT-5 Mini: 40.1 (#146), QwQ-32B: 35.4 (#226)
| Benchmark | GPT-5 Mini | QwQ-32B |
|---|---|---|
| LMArena Coding | 1406 | 1333 |
| SWE-bench Verified | 64.7% | — |
| SWE-bench Verified (bash only) | 59.8% | — |
| Aider Polyglot | — | 20.9% |
| SWE-bench Multilingual | 39.7% | — |
| SciCode | 39.2% | — |
| WeirdML | 52.7% | — |
| BigCodeBench Instruct | — | 44.6% |
| LiveBench Coding | — | 72.2% |
| BigCodeBench Complete | — | 54.4% |
| ALE-Bench | 799.77 | — |
| AlgoTune | 1.38 | — |
Agentic & Tool Use Not comparable
GPT-5 Mini: 31.1 (#70), QwQ-32B: —
| Benchmark | GPT-5 Mini | QwQ-32B |
|---|---|---|
| Terminal-Bench | 34.8% | — |
| Berkeley Function Calling Leaderboard | 55.5% | — |
| Vending-Bench 2 | -31.18 | — |
Reasoning Too close to call
GPT-5 Mini: 23.9 (#168), QwQ-32B: 23.7 (#174)
| Benchmark | GPT-5 Mini | QwQ-32B |
|---|---|---|
| Chess Puzzles | 30% | 5% |
| LMArena Hard Prompts | 1380 | 1325 |
| Epoch Capabilities Index | 145.52 | 137.6 |
| ForecastBench | 61 | 58.3 |
| ARC-AGI-2 | 4.4% | — |
| Kagi LLM Benchmark | 70.3% | — |
| ARC-AGI-1 | 54.3% | — |
| CritPt | 0% | — |
| EnigmaEval | 8.2% | — |
| LiveBench Reasoning | — | 83.5% |
| Mystery Game Puzzles | 10% | — |
| DTBench | 80.5% | — |
| LiveBench Data Analysis | — | 65% |
| LMCA | 34.2% | — |
| LiveBench | — | 72% |
Math GPT-5 Mini leads
GPT-5 Mini: 46.7 (#69), QwQ-32B: 38.0 (#143)
| Benchmark | GPT-5 Mini | QwQ-32B |
|---|---|---|
| OTIS Mock AIME 2024-2025 | 86.7% | 59.2% |
| LMArena Math | 1378 | 1359 |
| FrontierMath (Tiers 1-3) | 46.7% | — |
| FrontierMath Tier 4 | 12.2% | — |
| ProofBench | 9% | — |
| Omni-MATH | 72.2% | — |
| LiveBench Math | — | 77.8% |
| MATH Level 5 | 97.8% | — |
| FrontierMath (Feb 2025 set) | 27.2% | — |
| FrontierMath Tier 4 (v1) | 6.3% | — |
Knowledge GPT-5 Mini leads
GPT-5 Mini: 45.6 (#86), QwQ-32B: 37.2 (#158)
| Benchmark | GPT-5 Mini | QwQ-32B |
|---|---|---|
| GPQA Diamond | 75% | 65.3% |
| Confabulations | 13.3% | 15.6% |
| LMArena Expert | 1379 | 1324 |
| Humanity's Last Exam | 19.4% | — |
| SimpleQA Verified | 21.6% | — |
| MMLU-Pro | 83.5% | — |
| Vectara Hallucination Rate | 12.9% | — |
| GPQA (HELM) | 75.6% | — |
Multimodal Not comparable
GPT-5 Mini: 35.6 (#85), QwQ-32B: —
| Benchmark | GPT-5 Mini | QwQ-32B |
|---|---|---|
| LMArena Vision | 1202 | — |
| VPCT | 40.2% | — |
Multilingual GPT-5 Mini leads
GPT-5 Mini: 48.9 (#137), QwQ-32B: 44.8 (#176)
| Benchmark | GPT-5 Mini | QwQ-32B |
|---|---|---|
| LMArena Non-English | 1363 | 1305 |
| LMArena Chinese | 1385 | 1378 |
| LMArena French | 1386 | 1336 |
| LMArena German | 1366 | 1313 |
| LMArena Japanese | 1341 | 1262 |
| LMArena Korean | 1308 | 1279 |
| LMArena Russian | 1362 | 1297 |
| LMArena Spanish | 1355 | 1354 |
Instruction Following GPT-5 Mini leads
GPT-5 Mini: 76.2 (#46), QwQ-32B: 72.6 (#137)
| Benchmark | GPT-5 Mini | QwQ-32B |
|---|---|---|
| LMArena Instruction Following | 1357 | 1297 |
| LiveBench Instruction Following | — | 81.8% |
| IFEval | 92.7% | — |
Long Context QwQ-32B leads
GPT-5 Mini: 41.9 (#132), QwQ-32B: 49.0 (#11)
| Benchmark | GPT-5 Mini | QwQ-32B |
|---|---|---|
| Fiction.LiveBench | 69.4% | 83.3% |
| LMArena Longer Query | 1355 | 1308 |
Writing & Preference GPT-5 Mini leads
GPT-5 Mini: 55.2 (#148), QwQ-32B: 50.6 (#180)
| Benchmark | GPT-5 Mini | QwQ-32B |
|---|---|---|
| LMArena Text | 1373 | 1329 |
| LMArena Creative Writing | 1325 | 1288 |
| Short-Story Creative Writing | 83.1% | 80.2% |
| EQ-Bench Creative Writing | 1313 | 1257 |
| LMArena Multi-Turn | 1363 | 1314 |
| WildBench | 85.5% | — |
| LiveBench Language | — | 51.4% |
Frequently asked questions
Is GPT-5 Mini better than QwQ-32B?
GPT-5 Mini is the stronger model overall, scoring 41.8 to 39.8 on the Noometry Index.
Is GPT-5 Mini or QwQ-32B better for coding?
GPT-5 Mini scores higher on coding benchmarks: 40.1 versus 35.4 in the Noometry coding category.
How many benchmarks do GPT-5 Mini and QwQ-32B share?
26 benchmarks have published results for both models. GPT-5 Mini has 60 scored results on Noometry and QwQ-32B has 36.