Model comparison
GPT-5 Nano vs Qwen3.5 27B
Qwen3.5 27B is the stronger model overall, scoring 41.9 to 33.5 on the Noometry Index. GPT-5 Nano costs 6.0× less per token, which makes it the better buy when Qwen3.5 27B's lead doesn't matter for your workload.
Last verified . 22 shared benchmarks.
Alibaba (Qwen)Summary
- They share 22 benchmarks with published results for both. GPT-5 Nano scores higher in 1 category and Qwen3.5 27B in 8 categories; 9 gaps are clear of the uncertainty.
- The widest gap is in writing & preference, where Qwen3.5 27B leads 59.3 to 39.1.
- The biggest single-benchmark swing is LMCA: 7.9% for GPT-5 Nano and 34% for Qwen3.5 27B.
- GPT-5 Nano is cheaper at $0.05 / $0.40 per million input/output tokens, against $0.30 / $2.40 for Qwen3.5 27B.
- GPT-5 Nano accepts more context: 400K tokens versus 262K.
- Qwen3.5 27B has downloadable open weights; the other is API-only.
Side by side
| GPT-5 Nano | Qwen3.5 27B | |
|---|---|---|
| Provider | OpenAI | Alibaba (Qwen) |
| Noometry Index | 33.5 | 41.9 |
| Released | 2025-08-07 | 2026-02-23 |
| Weights | Proprietary | Open |
| Context window | 400K | 262K |
| Max output | 128K | 66K |
| Input $ / M tokens | $0.05 | $0.30 |
| Output $ / M tokens | $0.40 | $2.40 |
| Results tracked | 49 | 28 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding Qwen3.5 27B leads
GPT-5 Nano: 33.6 (#254), Qwen3.5 27B: 38.9 (#168)
| Benchmark | GPT-5 Nano | Qwen3.5 27B |
|---|---|---|
| WeirdML | 38.1% | 39.5% |
| LMArena Coding | 1351 | 1427 |
| ALE-Bench | 718.67 | 349.45 |
| SWE-bench Verified (bash only) | 34.8% | — |
| LMArena WebDev | — | 1358 |
Agentic & Tool Use Not comparable
GPT-5 Nano: 25.8 (#106), Qwen3.5 27B: —
| Benchmark | GPT-5 Nano | Qwen3.5 27B |
|---|---|---|
| Terminal-Bench | 21.8% | — |
| Berkeley Function Calling Leaderboard | 51.5% | — |
| Vending-Bench 2 | — | 201.98 |
Reasoning Qwen3.5 27B leads
GPT-5 Nano: 16.3 (#306), Qwen3.5 27B: 27.5 (#117)
| Benchmark | GPT-5 Nano | Qwen3.5 27B |
|---|---|---|
| LMArena Hard Prompts | 1328 | 1414 |
| DTBench | 62.7% | 82.4% |
| LMCA | 7.9% | 34% |
| ARC-AGI-2 | 2.6% | — |
| Kagi LLM Benchmark | 62.2% | — |
| NYT Connections (extended) | — | 47.9% |
| ARC-AGI-1 | 20.7% | — |
| Chess Puzzles | 27% | — |
| Thematic Generalization | — | 45.5% |
| Mystery Game Puzzles | 9% | — |
| Epoch Capabilities Index | 139.38 | — |
| ForecastBench | 59.1 | — |
Math Qwen3.5 27B leads
GPT-5 Nano: 29.4 (#241), Qwen3.5 27B: 38.8 (#127)
| Benchmark | GPT-5 Nano | Qwen3.5 27B |
|---|---|---|
| LMArena Math | 1317 | 1429 |
| FrontierMath (Tiers 1-3) | 20% | — |
| FrontierMath Tier 4 | 2.4% | — |
| MathArena Final-Answer Competitions | — | 56.7% |
| OTIS Mock AIME 2024-2025 | 81.1% | — |
| ProofBench | 12% | — |
| Omni-MATH | 54.6% | — |
| MATH Level 5 | 95.2% | — |
| FrontierMath (Feb 2025 set) | 8.3% | — |
| FrontierMath Tier 4 (v1) | 2.1% | — |
Knowledge Qwen3.5 27B leads
GPT-5 Nano: 35.9 (#178), Qwen3.5 27B: 38.0 (#150)
| Benchmark | GPT-5 Nano | Qwen3.5 27B |
|---|---|---|
| Vectara Hallucination Rate | 10.5% | 12.1% |
| LMArena Expert | 1321 | 1428 |
| GPQA Diamond | 69.4% | — |
| SimpleQA Verified | 11.7% | — |
| MMLU-Pro | 77.8% | — |
| GPQA (HELM) | 67.9% | — |
Multimodal Qwen3.5 27B leads
GPT-5 Nano: 31.3 (#108), Qwen3.5 27B: 39.4 (#59)
| Benchmark | GPT-5 Nano | Qwen3.5 27B |
|---|---|---|
| LMArena Vision | 1159 | 1241 |
| VPCT | 37.2% | — |
Multilingual Qwen3.5 27B leads
GPT-5 Nano: 45.3 (#172), Qwen3.5 27B: 50.8 (#115)
| Benchmark | GPT-5 Nano | Qwen3.5 27B |
|---|---|---|
| LMArena Non-English | 1313 | 1390 |
| LMArena Chinese | 1356 | 1478 |
| LMArena German | 1327 | 1393 |
| LMArena Japanese | 1226 | 1345 |
| LMArena Korean | 1269 | 1358 |
| LMArena Russian | 1296 | 1390 |
| LMArena Spanish | 1360 | 1407 |
| LMArena French | — | 1410 |
Instruction Following GPT-5 Nano leads
GPT-5 Nano: 75.0 (#79), Qwen3.5 27B: 73.5 (#119)
| Benchmark | GPT-5 Nano | Qwen3.5 27B |
|---|---|---|
| LMArena Instruction Following | 1306 | 1393 |
| IFEval | 93.2% | — |
Long Context Qwen3.5 27B leads
GPT-5 Nano: 31.3 (#281), Qwen3.5 27B: 43.1 (#106)
| Benchmark | GPT-5 Nano | Qwen3.5 27B |
|---|---|---|
| LMArena Longer Query | 1312 | 1413 |
| Fiction.LiveBench | 44.4% | — |
Writing & Preference Qwen3.5 27B leads
GPT-5 Nano: 39.1 (#249), Qwen3.5 27B: 59.3 (#111)
| Benchmark | GPT-5 Nano | Qwen3.5 27B |
|---|---|---|
| LMArena Text | 1320 | 1409 |
| LMArena Creative Writing | 1249 | 1362 |
| LMArena Multi-Turn | 1311 | 1410 |
| EQ-Bench Creative Writing | 705 | — |
| WildBench | 80.6% | — |
Frequently asked questions
Is GPT-5 Nano better than Qwen3.5 27B?
Qwen3.5 27B is the stronger model overall, scoring 41.9 to 33.5 on the Noometry Index. GPT-5 Nano costs 6.0× less per token, which makes it the better buy when Qwen3.5 27B's lead doesn't matter for your workload.
Which is cheaper, GPT-5 Nano or Qwen3.5 27B?
GPT-5 Nano is cheaper. It lists at $0.05 per million input tokens and $0.40 per million output tokens; Qwen3.5 27B lists at $0.30 and $2.40.
Is GPT-5 Nano or Qwen3.5 27B better for coding?
Qwen3.5 27B scores higher on coding benchmarks: 38.9 versus 33.6 in the Noometry coding category.
Which has the bigger context window?
GPT-5 Nano does, with 400K tokens against 262K.
How many benchmarks do GPT-5 Nano and Qwen3.5 27B share?
22 benchmarks have published results for both models. GPT-5 Nano has 49 scored results on Noometry and Qwen3.5 27B has 28.