Model comparison
GPT-3.5-turbo vs Qwen3.5 27B
Qwen3.5 27B is the stronger model overall, scoring 41.9 to 23.2 on the Noometry Index.
Last verified . 20 shared benchmarks.
Alibaba (Qwen)Summary
- They share 20 benchmarks with published results for both. GPT-3.5-turbo scores higher in 0 categories and Qwen3.5 27B in 8 categories; 8 gaps are clear of the uncertainty.
- The widest gap is in writing & preference, where Qwen3.5 27B leads 59.3 to 25.3.
- The biggest single-benchmark swing is WeirdML: 3.5% for GPT-3.5-turbo and 39.5% for Qwen3.5 27B.
- GPT-3.5-turbo is cheaper at $0.50 / $1.50 per million input/output tokens, against $0.30 / $2.40 for Qwen3.5 27B.
- Qwen3.5 27B accepts more context: 262K tokens versus 16K.
- Qwen3.5 27B has downloadable open weights; the other is API-only.
Side by side
| GPT-3.5-turbo | Qwen3.5 27B | |
|---|---|---|
| Provider | OpenAI | Alibaba (Qwen) |
| Noometry Index | 23.2 | 41.9 |
| Released | 2023-03-01 | 2026-02-23 |
| Weights | Proprietary | Open |
| Context window | 16K | 262K |
| Max output | 4K | 66K |
| Input $ / M tokens | $0.50 | $0.30 |
| Output $ / M tokens | $1.50 | $2.40 |
| Results tracked | 44 | 28 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding Qwen3.5 27B leads
GPT-3.5-turbo: 23.9 (#331), Qwen3.5 27B: 38.9 (#168)
| Benchmark | GPT-3.5-turbo | Qwen3.5 27B |
|---|---|---|
| WeirdML | 3.5% | 39.5% |
| LMArena Coding | 1136 | 1427 |
| LMArena WebDev | — | 1358 |
| BigCodeBench Instruct | 39.1% | — |
| BigCodeBench Complete | 50.6% | — |
| ALE-Bench | — | 349.45 |
| HumanEval+ | 70.7% | — |
| MBPP+ | 69.7% | — |
Agentic & Tool Use Not comparable
GPT-3.5-turbo: —, Qwen3.5 27B: —
| Benchmark | GPT-3.5-turbo | Qwen3.5 27B |
|---|---|---|
| METR Time Horizons | 21.5% | — |
| Vending-Bench 2 | — | 201.98 |
Reasoning Qwen3.5 27B leads
GPT-3.5-turbo: 13.8 (#332), Qwen3.5 27B: 27.5 (#117)
| Benchmark | GPT-3.5-turbo | Qwen3.5 27B |
|---|---|---|
| LMArena Hard Prompts | 1108 | 1414 |
| DTBench | 48.5% | 82.4% |
| LMCA | 9.7% | 34% |
| NYT Connections (extended) | — | 47.9% |
| Chess Puzzles | 0% | — |
| Thematic Generalization | — | 45.5% |
| Mystery Game Puzzles | 3% | — |
| Adversarial NLI | 58.1% | — |
| BIG-Bench Hard | 61.6% | — |
| CommonsenseQA 2.0 | 57% | — |
| Epoch Capabilities Index | 118.55 | — |
| ForecastBench | 50.4 | — |
| WinoGrande | 81.6% | — |
Math Qwen3.5 27B leads
GPT-3.5-turbo: 6.3 (#327), Qwen3.5 27B: 38.8 (#127)
| Benchmark | GPT-3.5-turbo | Qwen3.5 27B |
|---|---|---|
| LMArena Math | 1142 | 1429 |
| FrontierMath (Tiers 1-3) | 0% | — |
| MathArena Final-Answer Competitions | — | 56.7% |
| OTIS Mock AIME 2024-2025 | 2.2% | — |
| MATH Level 5 | 15.9% | — |
| GSM8K | 57.8% | — |
Knowledge Qwen3.5 27B leads
GPT-3.5-turbo: 10.0 (#303), Qwen3.5 27B: 38.0 (#150)
| Benchmark | GPT-3.5-turbo | Qwen3.5 27B |
|---|---|---|
| LMArena Expert | 1070 | 1428 |
| GPQA Diamond | 28% | — |
| Vectara Hallucination Rate | — | 12.1% |
| ARC (AI2) Challenge | 87.4% | — |
| BoolQ | 87% | — |
| MMLU | 71.4% | — |
| OpenBookQA | 86% | — |
| TriviaQA | 85.8% | — |
Multimodal Not comparable
GPT-3.5-turbo: —, Qwen3.5 27B: 39.4 (#59)
| Benchmark | GPT-3.5-turbo | Qwen3.5 27B |
|---|---|---|
| LMArena Vision | — | 1241 |
Multilingual Qwen3.5 27B leads
GPT-3.5-turbo: 31.5 (#258), Qwen3.5 27B: 50.8 (#115)
| Benchmark | GPT-3.5-turbo | Qwen3.5 27B |
|---|---|---|
| LMArena Non-English | 1108 | 1390 |
| LMArena Chinese | 1075 | 1478 |
| LMArena French | 1118 | 1410 |
| LMArena German | 1090 | 1393 |
| LMArena Japanese | 1043 | 1345 |
| LMArena Korean | 1019 | 1358 |
| LMArena Russian | 1123 | 1390 |
| LMArena Spanish | 1121 | 1407 |
Instruction Following Qwen3.5 27B leads
GPT-3.5-turbo: 57.9 (#262), Qwen3.5 27B: 73.5 (#119)
| Benchmark | GPT-3.5-turbo | Qwen3.5 27B |
|---|---|---|
| LMArena Instruction Following | 1119 | 1393 |
Long Context Qwen3.5 27B leads
GPT-3.5-turbo: 34.0 (#254), Qwen3.5 27B: 43.1 (#106)
| Benchmark | GPT-3.5-turbo | Qwen3.5 27B |
|---|---|---|
| LMArena Longer Query | 1121 | 1413 |
Writing & Preference Qwen3.5 27B leads
GPT-3.5-turbo: 25.3 (#305), Qwen3.5 27B: 59.3 (#111)
| Benchmark | GPT-3.5-turbo | Qwen3.5 27B |
|---|---|---|
| LMArena Text | 1125 | 1409 |
| LMArena Creative Writing | 1092 | 1362 |
| LMArena Multi-Turn | 1117 | 1410 |
| EQ-Bench Creative Writing | 451 | — |
Frequently asked questions
Is GPT-3.5-turbo better than Qwen3.5 27B?
Qwen3.5 27B is the stronger model overall, scoring 41.9 to 23.2 on the Noometry Index.
Which is cheaper, GPT-3.5-turbo or Qwen3.5 27B?
GPT-3.5-turbo is cheaper. It lists at $0.50 per million input tokens and $1.50 per million output tokens; Qwen3.5 27B lists at $0.30 and $2.40.
Is GPT-3.5-turbo or Qwen3.5 27B better for coding?
Qwen3.5 27B scores higher on coding benchmarks: 38.9 versus 23.9 in the Noometry coding category.
Which has the bigger context window?
Qwen3.5 27B does, with 262K tokens against 16K.
How many benchmarks do GPT-3.5-turbo and Qwen3.5 27B share?
20 benchmarks have published results for both models. GPT-3.5-turbo has 44 scored results on Noometry and Qwen3.5 27B has 28.