Model comparison
GPT-4o vs QwQ-32B
QwQ-32B is the stronger model overall, scoring 39.8 to 28.6 on the Noometry Index.
Last verified . 35 shared benchmarks.
Alibaba (Qwen)Summary
- They share 35 benchmarks with published results for both. GPT-4o scores higher in 1 category and QwQ-32B in 7 categories; 8 gaps are clear of the uncertainty.
- The widest gap is in math, where QwQ-32B leads 38.0 to 10.6.
- The biggest single-benchmark swing is OTIS Mock AIME 2024-2025: 6.4% for GPT-4o and 59.2% for QwQ-32B.
- QwQ-32B has downloadable open weights; the other is API-only.
Side by side
| GPT-4o | QwQ-32B | |
|---|---|---|
| Provider | OpenAI | Alibaba (Qwen) |
| Noometry Index | 28.6 | 39.8 |
| Released | 2024-05-13 | 2024-11-28 |
| Weights | Proprietary | Open |
| Context window | 128K | — |
| Max output | 16K | — |
| Input $ / M tokens | $2.50 | — |
| Output $ / M tokens | $10 | — |
| Results tracked | 72 | 36 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding QwQ-32B leads
GPT-4o: 24.8 (#328), QwQ-32B: 35.4 (#226)
| Benchmark | GPT-4o | QwQ-32B |
|---|---|---|
| Aider Polyglot | 45.3% | 20.9% |
| BigCodeBench Instruct | 51.1% | 44.6% |
| LiveBench Coding | 51.4% | 72.2% |
| LMArena Coding | 1297 | 1333 |
| BigCodeBench Complete | 61.1% | 54.4% |
| SWE-bench Verified | 31% | — |
| SWE-bench Verified (bash only) | 21.6% | — |
| GSO | 0% | — |
| WeirdML | 25.1% | — |
| CadEval | 26% | — |
| HumanEval+ | 87.2% | — |
| MBPP+ | 72.2% | — |
Agentic & Tool Use Not comparable
GPT-4o: 21.0 (#141), QwQ-32B: —
| Benchmark | GPT-4o | QwQ-32B |
|---|---|---|
| GDPval | 9.9% | — |
| TheAgentCompany | 8.6% | — |
| Cybench | 12.5% | — |
| BALROG | 32.3% | — |
| LMArena Search | 1006 | — |
| METR Time Horizons | 40.8% | — |
Reasoning QwQ-32B leads
GPT-4o: 9.4 (#343), QwQ-32B: 23.7 (#174)
| Benchmark | GPT-4o | QwQ-32B |
|---|---|---|
| Chess Puzzles | 13% | 5% |
| LiveBench Reasoning | 55.8% | 83.5% |
| LMArena Hard Prompts | 1281 | 1325 |
| LiveBench Data Analysis | 60.9% | 65% |
| Epoch Capabilities Index | 128.97 | 137.6 |
| ForecastBench | 57.7 | 58.3 |
| LiveBench | 55.3% | 72% |
| ARC-AGI-2 | 0% | — |
| SimpleBench | 17.8% | — |
| ARC-AGI-1 | 4.5% | — |
| CritPt | 0% | — |
| EnigmaEval | 0.8% | — |
| DTBench | 64.5% | — |
| LMCA | 16.6% | — |
Math QwQ-32B leads
GPT-4o: 10.6 (#312), QwQ-32B: 38.0 (#143)
| Benchmark | GPT-4o | QwQ-32B |
|---|---|---|
| OTIS Mock AIME 2024-2025 | 6.4% | 59.2% |
| LiveBench Math | 49.5% | 77.8% |
| LMArena Math | 1285 | 1359 |
| FrontierMath (Tiers 1-3) | 0.4% | — |
| Omni-MATH | 29.3% | — |
| MATH Level 5 | 53.3% | — |
| FrontierMath (Feb 2025 set) | 0.3% | — |
Knowledge QwQ-32B leads
GPT-4o: 28.8 (#242), QwQ-32B: 37.2 (#158)
| Benchmark | GPT-4o | QwQ-32B |
|---|---|---|
| GPQA Diamond | 49.2% | 65.3% |
| Confabulations | 15.3% | 15.6% |
| LMArena Expert | 1250 | 1324 |
| Humanity's Last Exam | 2.7% | — |
| SimpleQA Verified | 26% | — |
| MMLU-Pro | 71.3% | — |
| Vectara Hallucination Rate | 9.6% | — |
| GPQA (HELM) | 52% | — |
| MMLU | 88.1% | — |
Multimodal Not comparable
GPT-4o: 34.5 (#91), QwQ-32B: —
| Benchmark | GPT-4o | QwQ-32B |
|---|---|---|
| LMArena Vision | 1137 | — |
| Video-MME | 71.9% | — |
| GeoBench | 71% | — |
| VPCT | 40% | — |
| ScienceQA | 88.5% | — |
Multilingual QwQ-32B leads
GPT-4o: 43.2 (#186), QwQ-32B: 44.8 (#176)
| Benchmark | GPT-4o | QwQ-32B |
|---|---|---|
| LMArena Non-English | 1283 | 1305 |
| LMArena Chinese | 1277 | 1378 |
| LMArena French | 1304 | 1336 |
| LMArena German | 1282 | 1313 |
| LMArena Japanese | 1257 | 1262 |
| LMArena Korean | 1234 | 1279 |
| LMArena Russian | 1286 | 1297 |
| LMArena Spanish | 1292 | 1354 |
Instruction Following QwQ-32B leads
GPT-4o: 66.6 (#207), QwQ-32B: 72.6 (#137)
| Benchmark | GPT-4o | QwQ-32B |
|---|---|---|
| LiveBench Instruction Following | 68.6% | 81.8% |
| LMArena Instruction Following | 1278 | 1297 |
| IFEval | 81.7% | — |
Long Context QwQ-32B leads
GPT-4o: 39.4 (#179), QwQ-32B: 49.0 (#11)
| Benchmark | GPT-4o | QwQ-32B |
|---|---|---|
| Fiction.LiveBench | 66.7% | 83.3% |
| LMArena Longer Query | 1289 | 1308 |
Writing & Preference GPT-4o leads
GPT-4o: 52.6 (#166), QwQ-32B: 50.6 (#180)
| Benchmark | GPT-4o | QwQ-32B |
|---|---|---|
| LMArena Text | 1300 | 1329 |
| LMArena Creative Writing | 1292 | 1288 |
| Short-Story Creative Writing | 81.8% | 80.2% |
| LMArena Multi-Turn | 1302 | 1314 |
| LiveBench Language | 47.6% | 51.4% |
| EQ-Bench Creative Writing | — | 1257 |
| WildBench | 82.8% | — |
Frequently asked questions
Is GPT-4o better than QwQ-32B?
QwQ-32B is the stronger model overall, scoring 39.8 to 28.6 on the Noometry Index.
Is GPT-4o or QwQ-32B better for coding?
QwQ-32B scores higher on coding benchmarks: 35.4 versus 24.8 in the Noometry coding category.
How many benchmarks do GPT-4o and QwQ-32B share?
35 benchmarks have published results for both models. GPT-4o has 72 scored results on Noometry and QwQ-32B has 36.