Model comparison
GPT-3.5-turbo vs Qwen Max
Qwen Max is the stronger model overall, scoring 34.7 to 23.2 on the Noometry Index. GPT-3.5-turbo costs 3.7× less per token, which makes it the better buy when Qwen Max's lead doesn't matter for your workload.
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 Qwen Max in 8 categories; 8 gaps are clear of the uncertainty.
- The widest gap is in writing & preference, where Qwen Max leads 47.8 to 25.3.
- The biggest single-benchmark swing is MATH Level 5: 15.9% for GPT-3.5-turbo and 67.2% for Qwen Max.
- GPT-3.5-turbo is cheaper at $0.50 / $1.50 per million input/output tokens, against $1.60 / $6.40 for Qwen Max.
- Qwen Max accepts more context: 33K tokens versus 16K.
Side by side
| GPT-3.5-turbo | Qwen Max | |
|---|---|---|
| Provider | OpenAI | Alibaba (Qwen) |
| Noometry Index | 23.2 | 34.7 |
| Released | 2023-03-01 | 2024-04-03 |
| Weights | Proprietary | Proprietary |
| Context window | 16K | 33K |
| Max output | 4K | 8K |
| Input $ / M tokens | $0.50 | $1.60 |
| Output $ / M tokens | $1.50 | $6.40 |
| Results tracked | 44 | 23 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding Qwen Max leads
GPT-3.5-turbo: 23.9 (#331), Qwen Max: 30.7 (#292)
| Benchmark | GPT-3.5-turbo | Qwen Max |
|---|---|---|
| LMArena Coding | 1136 | 1288 |
| Aider Polyglot | — | 21.8% |
| WeirdML | 3.5% | — |
| BigCodeBench Instruct | 39.1% | — |
| BigCodeBench Complete | 50.6% | — |
| HumanEval+ | 70.7% | — |
| MBPP+ | 69.7% | — |
Agentic & Tool Use Not comparable
GPT-3.5-turbo: —, Qwen Max: —
| Benchmark | GPT-3.5-turbo | Qwen Max |
|---|---|---|
| METR Time Horizons | 21.5% | — |
Reasoning Qwen Max leads
GPT-3.5-turbo: 13.8 (#332), Qwen Max: 25.1 (#151)
| Benchmark | GPT-3.5-turbo | Qwen Max |
|---|---|---|
| LMArena Hard Prompts | 1108 | 1269 |
| Chess Puzzles | 0% | — |
| Mystery Game Puzzles | 3% | — |
| DTBench | 48.5% | — |
| LMCA | 9.7% | — |
| 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 Qwen Max leads
GPT-3.5-turbo: 6.3 (#327), Qwen Max: 22.3 (#276)
| Benchmark | GPT-3.5-turbo | Qwen Max |
|---|---|---|
| OTIS Mock AIME 2024-2025 | 2.2% | 16.1% |
| LMArena Math | 1142 | 1275 |
| MATH Level 5 | 15.9% | 67.2% |
| FrontierMath (Tiers 1-3) | 0% | — |
| FrontierMath (Feb 2025 set) | — | 1% |
| GSM8K | 57.8% | — |
Knowledge Qwen Max leads
GPT-3.5-turbo: 10.0 (#303), Qwen Max: 30.3 (#228)
| Benchmark | GPT-3.5-turbo | Qwen Max |
|---|---|---|
| GPQA Diamond | 28% | 56.1% |
| LMArena Expert | 1070 | 1248 |
| ARC (AI2) Challenge | 87.4% | — |
| BoolQ | 87% | — |
| MMLU | 71.4% | — |
| OpenBookQA | 86% | — |
| TriviaQA | 85.8% | — |
Multilingual Qwen Max leads
GPT-3.5-turbo: 31.5 (#258), Qwen Max: 41.8 (#202)
| Benchmark | GPT-3.5-turbo | Qwen Max |
|---|---|---|
| LMArena Non-English | 1108 | 1263 |
| LMArena Chinese | 1075 | 1254 |
| LMArena French | 1118 | 1330 |
| LMArena German | 1090 | 1254 |
| LMArena Japanese | 1043 | 1205 |
| LMArena Korean | 1019 | 1142 |
| LMArena Russian | 1123 | 1274 |
| LMArena Spanish | 1121 | 1290 |
Instruction Following Qwen Max leads
GPT-3.5-turbo: 57.9 (#262), Qwen Max: 66.5 (#208)
| Benchmark | GPT-3.5-turbo | Qwen Max |
|---|---|---|
| LMArena Instruction Following | 1119 | 1262 |
Long Context Qwen Max leads
GPT-3.5-turbo: 34.0 (#254), Qwen Max: 39.4 (#180)
| Benchmark | GPT-3.5-turbo | Qwen Max |
|---|---|---|
| LMArena Longer Query | 1121 | 1288 |
| Fiction.LiveBench | — | 66.7% |
Writing & Preference Qwen Max leads
GPT-3.5-turbo: 25.3 (#305), Qwen Max: 47.8 (#205)
| Benchmark | GPT-3.5-turbo | Qwen Max |
|---|---|---|
| LMArena Text | 1125 | 1282 |
| LMArena Creative Writing | 1092 | 1248 |
| LMArena Multi-Turn | 1117 | 1277 |
| EQ-Bench Creative Writing | 451 | — |
Frequently asked questions
Is GPT-3.5-turbo better than Qwen Max?
Qwen Max is the stronger model overall, scoring 34.7 to 23.2 on the Noometry Index. GPT-3.5-turbo costs 3.7× less per token, which makes it the better buy when Qwen Max's lead doesn't matter for your workload.
Which is cheaper, GPT-3.5-turbo or Qwen Max?
GPT-3.5-turbo is cheaper. It lists at $0.50 per million input tokens and $1.50 per million output tokens; Qwen Max lists at $1.60 and $6.40.
Is GPT-3.5-turbo or Qwen Max better for coding?
Qwen Max scores higher on coding benchmarks: 30.7 versus 23.9 in the Noometry coding category.
Which has the bigger context window?
Qwen Max does, with 33K tokens against 16K.
How many benchmarks do GPT-3.5-turbo and Qwen Max share?
20 benchmarks have published results for both models. GPT-3.5-turbo has 44 scored results on Noometry and Qwen Max has 23.