Model comparison
GPT-4o vs Qwen1.5-32B
Qwen1.5-32B is the stronger model overall, scoring 30.0 to 28.5 on the Noometry Index.
Last verified . 21 shared benchmarks.
Alibaba (Qwen)Summary
- They share 21 benchmarks with published results for both. GPT-4o scores higher in 5 categories and Qwen1.5-32B in 3 categories; 8 gaps are clear of the uncertainty.
- The widest gap is in math, where Qwen1.5-32B leads 30.5 to 10.2.
- The biggest single-benchmark swing is BigCodeBench Complete: 61.1% for GPT-4o and 42% for Qwen1.5-32B.
- Qwen1.5-32B has downloadable open weights; the other is API-only.
Side by side
| GPT-4o | Qwen1.5-32B | |
|---|---|---|
| Provider | OpenAI | Alibaba (Qwen) |
| Noometry Index | 28.5 | 30.0 |
| Released | 2024-05-13 | 2024-02-04 |
| Weights | Proprietary | Open |
| Context window | 128K | — |
| Max output | 16K | — |
| Input $ / M tokens | $2.50 | — |
| Output $ / M tokens | $10 | — |
| Results tracked | 72 | 21 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding Qwen1.5-32B leads
GPT-4o: 25.1 (#323), Qwen1.5-32B: 31.1 (#278)
| Benchmark | GPT-4o | Qwen1.5-32B |
|---|---|---|
| BigCodeBench Instruct | 51.1% | 32.3% |
| LMArena Coding | 1297 | 1155 |
| BigCodeBench Complete | 61.1% | 42% |
| SWE-bench Verified | 31% | — |
| SWE-bench Verified (bash only) | 21.6% | — |
| Aider Polyglot | 45.3% | — |
| GSO | 0% | — |
| WeirdML | 25.1% | — |
| LiveBench Coding | 51.4% | — |
| CadEval | 26% | — |
| HumanEval+ | 87.2% | — |
| MBPP+ | 72.2% | — |
Agentic & Tool Use Not comparable
GPT-4o: 20.8 (#133), Qwen1.5-32B: —
| Benchmark | GPT-4o | Qwen1.5-32B |
|---|---|---|
| GDPval | 9.9% | — |
| TheAgentCompany | 8.6% | — |
| Cybench | 12.5% | — |
| BALROG | 32.3% | — |
| LMArena Search | 1006 | — |
| METR Time Horizons | 40.8% | — |
Reasoning Qwen1.5-32B leads
GPT-4o: 9.4 (#338), Qwen1.5-32B: 20.9 (#210)
| Benchmark | GPT-4o | Qwen1.5-32B |
|---|---|---|
| LMArena Hard Prompts | 1281 | 1131 |
| ARC-AGI-2 | 0% | — |
| SimpleBench | 17.8% | — |
| ARC-AGI-1 | 4.5% | — |
| CritPt | 0% | — |
| Chess Puzzles | 13% | — |
| EnigmaEval | 0.8% | — |
| LiveBench Reasoning | 55.8% | — |
| DTBench | 64.5% | — |
| LiveBench Data Analysis | 60.9% | — |
| LMCA | 16.6% | — |
| Epoch Capabilities Index | 128.97 | — |
| ForecastBench | 57.7 | — |
| LiveBench | 55.3% | — |
Math Qwen1.5-32B leads
GPT-4o: 10.2 (#303), Qwen1.5-32B: 30.5 (#207)
| Benchmark | GPT-4o | Qwen1.5-32B |
|---|---|---|
| LMArena Math | 1285 | 1155 |
| FrontierMath (Tiers 1-3) | 0.4% | — |
| OTIS Mock AIME 2024-2025 | 6.4% | — |
| Omni-MATH | 29.3% | — |
| LiveBench Math | 49.5% | — |
| MATH Level 5 | 53.3% | — |
| FrontierMath (Feb 2025 set) | 0.3% | — |
Knowledge GPT-4o leads
GPT-4o: 28.8 (#235), Qwen1.5-32B: 13.5 (#291)
| Benchmark | GPT-4o | Qwen1.5-32B |
|---|---|---|
| GPQA Diamond | 49.2% | 30.7% |
| LMArena Expert | 1250 | 1126 |
| MMLU | 88.1% | 74.4% |
| Humanity's Last Exam | 2.7% | — |
| SimpleQA Verified | 26% | — |
| MMLU-Pro | 71.3% | — |
| Confabulations | 15.3% | — |
| Vectara Hallucination Rate | 9.6% | — |
| GPQA (HELM) | 52% | — |
Multimodal Not comparable
GPT-4o: 33.9 (#85), Qwen1.5-32B: —
| Benchmark | GPT-4o | Qwen1.5-32B |
|---|---|---|
| LMArena Vision | 1137 | — |
| Video-MME | 71.9% | — |
| GeoBench | 71% | — |
| VPCT | 40% | — |
| ScienceQA | 88.5% | — |
Multilingual GPT-4o leads
GPT-4o: 43.4 (#181), Qwen1.5-32B: 31.6 (#254)
| Benchmark | GPT-4o | Qwen1.5-32B |
|---|---|---|
| LMArena Non-English | 1283 | 1106 |
| LMArena Chinese | 1277 | 1176 |
| LMArena French | 1303 | 1100 |
| LMArena German | 1282 | 1058 |
| LMArena Japanese | 1257 | 1027 |
| LMArena Korean | 1234 | 1007 |
| LMArena Russian | 1286 | 1073 |
| LMArena Spanish | 1292 | 1089 |
Instruction Following GPT-4o leads
GPT-4o: 66.8 (#202), Qwen1.5-32B: 58.0 (#260)
| Benchmark | GPT-4o | Qwen1.5-32B |
|---|---|---|
| LMArena Instruction Following | 1278 | 1116 |
| LiveBench Instruction Following | 68.6% | — |
| IFEval | 81.7% | — |
Long Context GPT-4o leads
GPT-4o: 39.4 (#174), Qwen1.5-32B: 34.7 (#241)
| Benchmark | GPT-4o | Qwen1.5-32B |
|---|---|---|
| LMArena Longer Query | 1289 | 1146 |
| Fiction.LiveBench | 66.7% | — |
Writing & Preference GPT-4o leads
GPT-4o: 52.8 (#160), Qwen1.5-32B: 34.4 (#266)
| Benchmark | GPT-4o | Qwen1.5-32B |
|---|---|---|
| LMArena Text | 1300 | 1137 |
| LMArena Creative Writing | 1291 | 1083 |
| LMArena Multi-Turn | 1302 | 1140 |
| Short-Story Creative Writing | 81.8% | — |
| WildBench | 82.8% | — |
| LiveBench Language | 47.6% | — |
Frequently asked questions
Is GPT-4o better than Qwen1.5-32B?
Qwen1.5-32B is the stronger model overall, scoring 30.0 to 28.5 on the Noometry Index.
Is GPT-4o or Qwen1.5-32B better for coding?
Qwen1.5-32B scores higher on coding benchmarks: 31.1 versus 25.1 in the Noometry coding category.
How many benchmarks do GPT-4o and Qwen1.5-32B share?
21 benchmarks have published results for both models. GPT-4o has 72 scored results on Noometry and Qwen1.5-32B has 21.