Model comparison
GPT-5.5 vs Qwen3-4B
GPT-5.5 is the stronger model overall, scoring 63.4 to 31.9 on the Noometry Index.
Last verified . 5 shared benchmarks.
Alibaba (Qwen)Summary
- They share 5 benchmarks with published results for both. GPT-5.5 scores higher in 4 categories and Qwen3-4B in 0 categories; 4 gaps are clear of the uncertainty.
- The widest gap is in reasoning, where GPT-5.5 leads 72.8 to 19.2.
- The biggest single-benchmark swing is MathArena Final-Answer Competitions: 94.3% for GPT-5.5 and 38.5% for Qwen3-4B.
- Qwen3-4B has downloadable open weights; the other is API-only.
Side by side
| GPT-5.5 | Qwen3-4B | |
|---|---|---|
| Provider | OpenAI | Alibaba (Qwen) |
| Noometry Index | 63.4 | 31.9 |
| Released | 2026-04-23 | 2025-04-29 |
| Weights | Proprietary | Open |
| Context window | 1.05M | — |
| Max output | 128K | — |
| Input $ / M tokens | $5 | — |
| Output $ / M tokens | $30 | — |
| Results tracked | 71 | 6 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding Not comparable
GPT-5.5: 58.2 (#17), Qwen3-4B: —
| Benchmark | GPT-5.5 | Qwen3-4B |
|---|---|---|
| SWE-bench Verified | 80.6% | — |
| DeepSWE | 67% | — |
| FrontierCode | 43% | — |
| LMArena WebDev | 1513 | — |
| SciCode | 56.1% | — |
| GSO | 40.2% | — |
| WeirdML | 84.9% | — |
| LMArena Coding | 1494 | — |
| MirrorCode | 10% | — |
| ALE-Bench | 1,943 | — |
Agentic & Tool Use GPT-5.5 leads
GPT-5.5: 50.7 (#6), Qwen3-4B: 27.6 (#100)
| Benchmark | GPT-5.5 | Qwen3-4B |
|---|---|---|
| Terminal-Bench | 84.7% | — |
| APEX-Agents | 55.1% | — |
| Berkeley Function Calling Leaderboard | — | 35.7% |
| OSWorld 2.0 | 13% | — |
| Remote Labor Index | 6.3% | — |
| τ²-bench Banking | 44.6% | — |
| DeepResearch Bench | 54% | — |
| PostTrainBench | 27.2% | — |
| ExploitBench | 47.4% | — |
| GBAEval | 53.2% | — |
| GDP.pdf | 26% | — |
| LMArena Search | 1242 | — |
| Vending-Bench 2 | 7,524 | — |
Reasoning GPT-5.5 leads
GPT-5.5: 72.8 (#11), Qwen3-4B: 19.2 (#268)
| Benchmark | GPT-5.5 | Qwen3-4B |
|---|---|---|
| Chess Puzzles | 54% | 4% |
| ARC-AGI-2 | 85% | — |
| SimpleBench | 69% | — |
| Kagi LLM Benchmark | 88.8% | — |
| NYT Connections (extended) | 96.2% | — |
| ARC-AGI-1 | 95% | — |
| CritPt | 27.1% | — |
| EBR-Bench | 34.3% | — |
| LMArena Hard Prompts | 1489 | — |
| Mystery Game Puzzles | 56% | — |
| DTBench | 96% | — |
| LMCA | 54.3% | — |
| Surface Evolver Bench | 88.1% | — |
| Bench to the Future 3 | 0.14 | — |
| Epoch Capabilities Index | 159.1 | — |
| ForecastBench | 60.6 | — |
Math GPT-5.5 leads
GPT-5.5: 81.7 (#11), Qwen3-4B: 29.7 (#240)
| Benchmark | GPT-5.5 | Qwen3-4B |
|---|---|---|
| MathArena Final-Answer Competitions | 94.3% | 38.5% |
| OTIS Mock AIME 2024-2025 | 100% | 52.2% |
| FrontierMath (Tiers 1-3) | 85.3% | — |
| FrontierMath Tier 4 | 72.5% | — |
| ProofBench | 50% | — |
| LMArena Math | 1486 | — |
| FrontierMath (Feb 2025 set) | 51.7% | — |
| FrontierMath Erdős | 0% | — |
| FrontierMath Tier 4 (v1) | 35.4% | — |
Knowledge GPT-5.5 leads
GPT-5.5: 64.4 (#17), Qwen3-4B: 33.0 (#208)
| Benchmark | GPT-5.5 | Qwen3-4B |
|---|---|---|
| GPQA Diamond | 94% | 52.3% |
| Vectara Hallucination Rate | 9.3% | 5.7% |
| SimpleQA Verified | 63% | — |
| LMArena Expert | 1508 | — |
Multimodal Not comparable
GPT-5.5: 46.9 (#12), Qwen3-4B: —
| Benchmark | GPT-5.5 | Qwen3-4B |
|---|---|---|
| LMArena Vision | 1297 | — |
| Blueprint-Bench 2 | 36.2% | — |
| Furniture Assembly | 44.2% | — |
| LMArena Document | 1486 | — |
Multilingual Not comparable
GPT-5.5: 56.4 (#20), Qwen3-4B: —
| Benchmark | GPT-5.5 | Qwen3-4B |
|---|---|---|
| LMArena Non-English | 1467 | — |
| LMArena Chinese | 1533 | — |
| LMArena French | 1486 | — |
| LMArena German | 1480 | — |
| LMArena Japanese | 1498 | — |
| LMArena Korean | 1460 | — |
| LMArena Russian | 1473 | — |
| LMArena Spanish | 1468 | — |
Instruction Following Not comparable
GPT-5.5: 77.5 (#18), Qwen3-4B: —
| Benchmark | GPT-5.5 | Qwen3-4B |
|---|---|---|
| LMArena Instruction Following | 1479 | — |
Long Context Not comparable
GPT-5.5: 48.3 (#12), Qwen3-4B: —
| Benchmark | GPT-5.5 | Qwen3-4B |
|---|---|---|
| CL-bench Life | 22.2% | — |
| LMArena Longer Query | 1484 | — |
Writing & Preference Not comparable
GPT-5.5: 72.7 (#13), Qwen3-4B: —
| Benchmark | GPT-5.5 | Qwen3-4B |
|---|---|---|
| LMArena Text | 1472 | — |
| LMArena Creative Writing | 1455 | — |
| EQ-Bench Creative Writing | 1844 | — |
| EQ-Bench 4 | 1315 | — |
| LMArena Multi-Turn | 1476 | — |
Frequently asked questions
Is GPT-5.5 better than Qwen3-4B?
GPT-5.5 is the stronger model overall, scoring 63.4 to 31.9 on the Noometry Index.
How many benchmarks do GPT-5.5 and Qwen3-4B share?
5 benchmarks have published results for both models. GPT-5.5 has 71 scored results on Noometry and Qwen3-4B has 6.