Model comparison
GPT-5.5 vs Qwen3.7 Max
GPT-5.5 is the stronger model overall, scoring 63.4 to 51.4 on the Noometry Index. Qwen3.7 Max costs 3.0× less per token, which makes it the better buy when GPT-5.5's lead doesn't matter for your workload.
Last verified . 33 shared benchmarks.
Alibaba (Qwen)Summary
- They share 33 benchmarks with published results for both. GPT-5.5 scores higher in 8 categories and Qwen3.7 Max in 1 category; 7 gaps are clear of the uncertainty.
- The widest gap is in agentic & tool use, where GPT-5.5 leads 50.7 to 21.8.
- The biggest single-benchmark swing is GBAEval: 53.2% for GPT-5.5 and 0.4% for Qwen3.7 Max.
- Qwen3.7 Max is cheaper at $2.50 / $7.50 per million input/output tokens, against $5 / $30 for GPT-5.5.
- GPT-5.5 accepts more context: 1.05M tokens versus 1M.
Side by side
| GPT-5.5 | Qwen3.7 Max | |
|---|---|---|
| Provider | OpenAI | Alibaba (Qwen) |
| Noometry Index | 63.4 | 51.4 |
| Released | 2026-04-23 | 2026-05-19 |
| Weights | Proprietary | Proprietary |
| Context window | 1.05M | 1M |
| Max output | 128K | 131K |
| Input $ / M tokens | $5 | $2.50 |
| Output $ / M tokens | $30 | $7.50 |
| Results tracked | 71 | 33 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding GPT-5.5 leads
GPT-5.5: 58.3 (#13), Qwen3.7 Max: 50.2 (#43)
| Benchmark | GPT-5.5 | Qwen3.7 Max |
|---|---|---|
| SWE-bench Verified | 80.6% | 77.3% |
| LMArena WebDev | 1512 | 1515 |
| SciCode | 56.1% | 48.8% |
| LMArena Coding | 1494 | 1498 |
| ALE-Bench | 1,943 | 1,189 |
| DeepSWE | 67% | — |
| FrontierCode | 43% | — |
| GSO | 40.2% | — |
| WeirdML | 84.9% | — |
| MirrorCode | 10% | — |
Agentic & Tool Use GPT-5.5 leads
GPT-5.5: 50.7 (#6), Qwen3.7 Max: 21.8 (#129)
| Benchmark | GPT-5.5 | Qwen3.7 Max |
|---|---|---|
| GBAEval | 53.2% | 0.4% |
| Terminal-Bench | 84.7% | — |
| APEX-Agents | 55.1% | — |
| OSWorld 2.0 | 13% | — |
| Remote Labor Index | 6.3% | — |
| τ²-bench Banking | 44.6% | — |
| DeepResearch Bench | 54% | — |
| PostTrainBench | 27.2% | — |
| ExploitBench | 47.4% | — |
| GDP.pdf | 26% | — |
| LMArena Search | 1242 | — |
| Vending-Bench 2 | 7,524 | — |
Reasoning GPT-5.5 leads
GPT-5.5: 72.7 (#8), Qwen3.7 Max: 48.4 (#36)
| Benchmark | GPT-5.5 | Qwen3.7 Max |
|---|---|---|
| SimpleBench | 69% | 70.4% |
| NYT Connections (extended) | 96.2% | 85.1% |
| CritPt | 27.1% | 13.4% |
| Chess Puzzles | 54% | 19% |
| EBR-Bench | 34.3% | 9.5% |
| LMArena Hard Prompts | 1489 | 1483 |
| Mystery Game Puzzles | 56% | 32% |
| DTBench | 96% | 92.3% |
| LMCA | 54.3% | 44% |
| Epoch Capabilities Index | 159.25 | 153.74 |
| ARC-AGI-2 | 85% | — |
| Kagi LLM Benchmark | 88.8% | — |
| ARC-AGI-1 | 95% | — |
| Surface Evolver Bench | 88.1% | — |
| Bench to the Future 3 | 0.14 | — |
| ForecastBench | 60.6 | — |
Math GPT-5.5 leads
GPT-5.5: 81.5 (#9), Qwen3.7 Max: 62.3 (#26)
| Benchmark | GPT-5.5 | Qwen3.7 Max |
|---|---|---|
| FrontierMath (Tiers 1-3) | 85.3% | 64.6% |
| FrontierMath Tier 4 | 72.5% | 34.1% |
| OTIS Mock AIME 2024-2025 | 100% | 95.6% |
| ProofBench | 50% | 26% |
| LMArena Math | 1486 | 1488 |
| MathArena Final-Answer Competitions | 94.3% | — |
| 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 (#14), Qwen3.7 Max: 61.7 (#24)
| Benchmark | GPT-5.5 | Qwen3.7 Max |
|---|---|---|
| GPQA Diamond | 94% | 90.9% |
| SimpleQA Verified | 63% | 55.8% |
| LMArena Expert | 1508 | 1488 |
| Vectara Hallucination Rate | 9.3% | — |
Multimodal Not comparable
GPT-5.5: 47.1 (#11), Qwen3.7 Max: —
| Benchmark | GPT-5.5 | Qwen3.7 Max |
|---|---|---|
| LMArena Vision | 1297 | — |
| Blueprint-Bench 2 | 36.2% | — |
| Furniture Assembly | 44.2% | — |
| LMArena Document | 1486 | — |
Multilingual Too close to call
GPT-5.5: 56.5 (#19), Qwen3.7 Max: 57.1 (#13)
| Benchmark | GPT-5.5 | Qwen3.7 Max |
|---|---|---|
| LMArena Non-English | 1467 | 1475 |
| LMArena Chinese | 1535 | 1533 |
| LMArena Russian | 1473 | 1482 |
| LMArena French | 1486 | — |
| LMArena German | 1478 | — |
| LMArena Japanese | 1498 | — |
| LMArena Korean | 1461 | — |
| LMArena Spanish | 1469 | — |
Instruction Following Too close to call
GPT-5.5: 77.7 (#18), Qwen3.7 Max: 76.9 (#34)
| Benchmark | GPT-5.5 | Qwen3.7 Max |
|---|---|---|
| LMArena Instruction Following | 1479 | 1460 |
Long Context GPT-5.5 leads
GPT-5.5: 48.3 (#12), Qwen3.7 Max: 45.4 (#34)
| Benchmark | GPT-5.5 | Qwen3.7 Max |
|---|---|---|
| LMArena Longer Query | 1484 | 1484 |
| CL-bench Life | 22.2% | — |
Writing & Preference GPT-5.5 leads
GPT-5.5: 72.8 (#13), Qwen3.7 Max: 65.2 (#52)
| Benchmark | GPT-5.5 | Qwen3.7 Max |
|---|---|---|
| LMArena Text | 1471 | 1476 |
| LMArena Creative Writing | 1455 | 1450 |
| EQ-Bench 4 | 1315 | 1110 |
| LMArena Multi-Turn | 1477 | 1482 |
| EQ-Bench Creative Writing | 1844 | — |
Frequently asked questions
Is GPT-5.5 better than Qwen3.7 Max?
GPT-5.5 is the stronger model overall, scoring 63.4 to 51.4 on the Noometry Index. Qwen3.7 Max costs 3.0× less per token, which makes it the better buy when GPT-5.5's lead doesn't matter for your workload.
Which is cheaper, GPT-5.5 or Qwen3.7 Max?
Qwen3.7 Max is cheaper. It lists at $2.50 per million input tokens and $7.50 per million output tokens; GPT-5.5 lists at $5 and $30.
Is GPT-5.5 or Qwen3.7 Max better for coding?
GPT-5.5 scores higher on coding benchmarks: 58.3 versus 50.2 in the Noometry coding category.
Which has the bigger context window?
GPT-5.5 does, with 1.05M tokens against 1M.
How many benchmarks do GPT-5.5 and Qwen3.7 Max share?
33 benchmarks have published results for both models. GPT-5.5 has 71 scored results on Noometry and Qwen3.7 Max has 33.