Model comparison
GPT-5 Nano vs Qwen3.7 Max
Qwen3.7 Max is the stronger model overall, scoring 51.4 to 33.4 on the Noometry Index. GPT-5 Nano costs 27× less per token, which makes it the better buy when Qwen3.7 Max's lead doesn't matter for your workload.
Last verified . 24 shared benchmarks.
Alibaba (Qwen)Summary
- They share 24 benchmarks with published results for both. GPT-5 Nano scores higher in 1 category and Qwen3.7 Max in 8 categories; 9 gaps are clear of the uncertainty.
- The widest gap is in math, where Qwen3.7 Max leads 62.3 to 29.1.
- The biggest single-benchmark swing is FrontierMath (Tiers 1-3): 20% for GPT-5 Nano and 64.6% for Qwen3.7 Max.
- GPT-5 Nano is cheaper at $0.05 / $0.40 per million input/output tokens, against $2.50 / $7.50 for Qwen3.7 Max.
- Qwen3.7 Max accepts more context: 1M tokens versus 400K.
Side by side
| GPT-5 Nano | Qwen3.7 Max | |
|---|---|---|
| Provider | OpenAI | Alibaba (Qwen) |
| Noometry Index | 33.4 | 51.4 |
| Released | 2025-08-07 | 2026-05-19 |
| Weights | Proprietary | Proprietary |
| Context window | 400K | 1M |
| Max output | 128K | 131K |
| Input $ / M tokens | $0.05 | $2.50 |
| Output $ / M tokens | $0.40 | $7.50 |
| Results tracked | 48 | 33 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding Qwen3.7 Max leads
GPT-5 Nano: 33.3 (#248), Qwen3.7 Max: 50.2 (#43)
| Benchmark | GPT-5 Nano | Qwen3.7 Max |
|---|---|---|
| LMArena Coding | 1351 | 1498 |
| ALE-Bench | 718.67 | 1,189 |
| SWE-bench Verified | — | 77.3% |
| SWE-bench Verified (bash only) | 34.8% | — |
| LMArena WebDev | — | 1515 |
| SciCode | — | 48.8% |
| WeirdML | 38.1% | — |
Agentic & Tool Use GPT-5 Nano leads
GPT-5 Nano: 25.6 (#99), Qwen3.7 Max: 21.8 (#129)
| Benchmark | GPT-5 Nano | Qwen3.7 Max |
|---|---|---|
| Terminal-Bench | 21.8% | — |
| Berkeley Function Calling Leaderboard | 51.5% | — |
| GBAEval | — | 0.4% |
Reasoning Qwen3.7 Max leads
GPT-5 Nano: 16.4 (#291), Qwen3.7 Max: 48.4 (#36)
| Benchmark | GPT-5 Nano | Qwen3.7 Max |
|---|---|---|
| Chess Puzzles | 27% | 19% |
| LMArena Hard Prompts | 1329 | 1483 |
| Mystery Game Puzzles | 9% | 32% |
| DTBench | 62.7% | 92.3% |
| LMCA | 7.9% | 44% |
| Epoch Capabilities Index | 139.39 | 153.74 |
| ARC-AGI-2 | 2.6% | — |
| SimpleBench | — | 70.4% |
| Kagi LLM Benchmark | 62.2% | — |
| NYT Connections (extended) | — | 85.1% |
| ARC-AGI-1 | 20.7% | — |
| CritPt | — | 13.4% |
| EBR-Bench | — | 9.5% |
| ForecastBench | 59.1 | — |
Math Qwen3.7 Max leads
GPT-5 Nano: 29.1 (#221), Qwen3.7 Max: 62.3 (#26)
| Benchmark | GPT-5 Nano | Qwen3.7 Max |
|---|---|---|
| FrontierMath (Tiers 1-3) | 20% | 64.6% |
| FrontierMath Tier 4 | 2.4% | 34.1% |
| OTIS Mock AIME 2024-2025 | 81.1% | 95.6% |
| ProofBench | 12% | 26% |
| LMArena Math | 1318 | 1488 |
| Omni-MATH | 54.6% | — |
| MATH Level 5 | 95.2% | — |
| FrontierMath (Feb 2025 set) | 8.3% | — |
| FrontierMath Tier 4 (v1) | 2.1% | — |
Knowledge Qwen3.7 Max leads
GPT-5 Nano: 35.8 (#166), Qwen3.7 Max: 61.7 (#24)
| Benchmark | GPT-5 Nano | Qwen3.7 Max |
|---|---|---|
| GPQA Diamond | 69.4% | 90.9% |
| SimpleQA Verified | 11.7% | 55.8% |
| LMArena Expert | 1321 | 1488 |
| MMLU-Pro | 77.8% | — |
| Vectara Hallucination Rate | 10.5% | — |
| GPQA (HELM) | 67.9% | — |
Multimodal Not comparable
GPT-5 Nano: 30.7 (#102), Qwen3.7 Max: —
| Benchmark | GPT-5 Nano | Qwen3.7 Max |
|---|---|---|
| LMArena Vision | 1159 | — |
| VPCT | 37.2% | — |
Multilingual Qwen3.7 Max leads
GPT-5 Nano: 45.5 (#167), Qwen3.7 Max: 57.1 (#13)
| Benchmark | GPT-5 Nano | Qwen3.7 Max |
|---|---|---|
| LMArena Non-English | 1313 | 1475 |
| LMArena Chinese | 1355 | 1533 |
| LMArena Russian | 1296 | 1482 |
| LMArena German | 1329 | — |
| LMArena Korean | 1269 | — |
| LMArena Spanish | 1360 | — |
Instruction Following Qwen3.7 Max leads
GPT-5 Nano: 75.2 (#72), Qwen3.7 Max: 76.9 (#34)
| Benchmark | GPT-5 Nano | Qwen3.7 Max |
|---|---|---|
| LMArena Instruction Following | 1306 | 1460 |
| IFEval | 93.2% | — |
Long Context Qwen3.7 Max leads
GPT-5 Nano: 31.3 (#275), Qwen3.7 Max: 45.4 (#34)
| Benchmark | GPT-5 Nano | Qwen3.7 Max |
|---|---|---|
| LMArena Longer Query | 1313 | 1484 |
| Fiction.LiveBench | 44.4% | — |
Writing & Preference Qwen3.7 Max leads
GPT-5 Nano: 39.4 (#244), Qwen3.7 Max: 65.2 (#52)
| Benchmark | GPT-5 Nano | Qwen3.7 Max |
|---|---|---|
| LMArena Text | 1320 | 1476 |
| LMArena Creative Writing | 1250 | 1450 |
| LMArena Multi-Turn | 1311 | 1482 |
| EQ-Bench Creative Writing | 705 | — |
| WildBench | 80.6% | — |
| EQ-Bench 4 | — | 1110 |
Frequently asked questions
Is GPT-5 Nano better than Qwen3.7 Max?
Qwen3.7 Max is the stronger model overall, scoring 51.4 to 33.4 on the Noometry Index. GPT-5 Nano costs 27× less per token, which makes it the better buy when Qwen3.7 Max's lead doesn't matter for your workload.
Which is cheaper, GPT-5 Nano or Qwen3.7 Max?
GPT-5 Nano is cheaper. It lists at $0.05 per million input tokens and $0.40 per million output tokens; Qwen3.7 Max lists at $2.50 and $7.50.
Is GPT-5 Nano or Qwen3.7 Max better for coding?
Qwen3.7 Max scores higher on coding benchmarks: 50.2 versus 33.3 in the Noometry coding category.
Which has the bigger context window?
Qwen3.7 Max does, with 1M tokens against 400K.
How many benchmarks do GPT-5 Nano and Qwen3.7 Max share?
24 benchmarks have published results for both models. GPT-5 Nano has 48 scored results on Noometry and Qwen3.7 Max has 33.