Model comparison
GPT-3.5-turbo vs Qwen3.8 Max
Qwen3.8 Max is the stronger model overall, scoring 56.8 to 23.2 on the Noometry Index. GPT-3.5-turbo costs 4.0× less per token, which makes it the better buy when Qwen3.8 Max's lead doesn't matter for your workload.
Last verified . 25 shared benchmarks.
Alibaba (Qwen)Summary
- They share 25 benchmarks with published results for both. GPT-3.5-turbo scores higher in 0 categories and Qwen3.8 Max in 8 categories; 8 gaps are clear of the uncertainty.
- The widest gap is in math, where Qwen3.8 Max leads 73.2 to 6.3.
- The biggest single-benchmark swing is OTIS Mock AIME 2024-2025: 2.2% for GPT-3.5-turbo and 100% for Qwen3.8 Max.
- GPT-3.5-turbo is cheaper at $0.50 / $1.50 per million input/output tokens, against $2 / $6 for Qwen3.8 Max.
- Qwen3.8 Max accepts more context: 1M tokens versus 16K.
Side by side
| GPT-3.5-turbo | Qwen3.8 Max | |
|---|---|---|
| Provider | OpenAI | Alibaba (Qwen) |
| Noometry Index | 23.2 | 56.8 |
| Released | 2023-03-01 | 2026-08-02 |
| Weights | Proprietary | Proprietary |
| Context window | 16K | 1M |
| Max output | 4K | 131K |
| Input $ / M tokens | $0.50 | $2 |
| Output $ / M tokens | $1.50 | $6 |
| Results tracked | 44 | 39 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding Qwen3.8 Max leads
GPT-3.5-turbo: 23.9 (#331), Qwen3.8 Max: 53.5 (#29)
| Benchmark | GPT-3.5-turbo | Qwen3.8 Max |
|---|---|---|
| LMArena Coding | 1136 | 1502 |
| DeepSWE | — | 57.5% |
| LMArena WebDev | — | 1674 |
| FrontierSWE | — | 17.8% |
| SciCode | — | 53.2% |
| 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: —, Qwen3.8 Max: 45.4 (#14)
| Benchmark | GPT-3.5-turbo | Qwen3.8 Max |
|---|---|---|
| APEX-Agents | — | 63.3% |
| τ²-bench Banking | — | 55.1% |
| GDP.pdf | — | 23.2% |
| METR Time Horizons | 21.5% | — |
Reasoning Qwen3.8 Max leads
GPT-3.5-turbo: 13.8 (#332), Qwen3.8 Max: 54.4 (#26)
| Benchmark | GPT-3.5-turbo | Qwen3.8 Max |
|---|---|---|
| Chess Puzzles | 0% | 40% |
| LMArena Hard Prompts | 1108 | 1496 |
| Mystery Game Puzzles | 3% | 38% |
| DTBench | 48.5% | 92% |
| LMCA | 9.7% | 46.2% |
| Epoch Capabilities Index | 118.55 | 156.41 |
| NYT Connections (extended) | — | 88.3% |
| CritPt | — | 20% |
| Adversarial NLI | 58.1% | — |
| BIG-Bench Hard | 61.6% | — |
| CommonsenseQA 2.0 | 57% | — |
| ForecastBench | 50.4 | — |
| WinoGrande | 81.6% | — |
Math Qwen3.8 Max leads
GPT-3.5-turbo: 6.3 (#327), Qwen3.8 Max: 73.2 (#20)
| Benchmark | GPT-3.5-turbo | Qwen3.8 Max |
|---|---|---|
| FrontierMath (Tiers 1-3) | 0% | 74.7% |
| OTIS Mock AIME 2024-2025 | 2.2% | 100% |
| LMArena Math | 1142 | 1499 |
| FrontierMath Tier 4 | — | 46.3% |
| ProofBench | — | 58% |
| MATH Level 5 | 15.9% | — |
| GSM8K | 57.8% | — |
Knowledge Qwen3.8 Max leads
GPT-3.5-turbo: 10.0 (#303), Qwen3.8 Max: 61.7 (#27)
| Benchmark | GPT-3.5-turbo | Qwen3.8 Max |
|---|---|---|
| GPQA Diamond | 28% | 92.7% |
| LMArena Expert | 1070 | 1507 |
| SimpleQA Verified | — | 47.3% |
| ARC (AI2) Challenge | 87.4% | — |
| BoolQ | 87% | — |
| MMLU | 71.4% | — |
| OpenBookQA | 86% | — |
| TriviaQA | 85.8% | — |
Multimodal Not comparable
GPT-3.5-turbo: —, Qwen3.8 Max: 37.2 (#75)
| Benchmark | GPT-3.5-turbo | Qwen3.8 Max |
|---|---|---|
| LMArena Vision | — | 1314 |
| Furniture Assembly | — | 20% |
Multilingual Qwen3.8 Max leads
GPT-3.5-turbo: 31.5 (#258), Qwen3.8 Max: 56.7 (#18)
| Benchmark | GPT-3.5-turbo | Qwen3.8 Max |
|---|---|---|
| LMArena Non-English | 1108 | 1472 |
| LMArena Chinese | 1075 | 1538 |
| LMArena French | 1118 | 1503 |
| LMArena German | 1090 | 1483 |
| LMArena Japanese | 1043 | 1467 |
| LMArena Korean | 1019 | 1461 |
| LMArena Russian | 1123 | 1481 |
| LMArena Spanish | 1121 | 1492 |
Instruction Following Qwen3.8 Max leads
GPT-3.5-turbo: 57.9 (#262), Qwen3.8 Max: 77.6 (#17)
| Benchmark | GPT-3.5-turbo | Qwen3.8 Max |
|---|---|---|
| LMArena Instruction Following | 1119 | 1479 |
Long Context Qwen3.8 Max leads
GPT-3.5-turbo: 34.0 (#254), Qwen3.8 Max: 45.6 (#31)
| Benchmark | GPT-3.5-turbo | Qwen3.8 Max |
|---|---|---|
| LMArena Longer Query | 1121 | 1489 |
Writing & Preference Qwen3.8 Max leads
GPT-3.5-turbo: 25.3 (#305), Qwen3.8 Max: 67.1 (#30)
| Benchmark | GPT-3.5-turbo | Qwen3.8 Max |
|---|---|---|
| LMArena Text | 1125 | 1483 |
| LMArena Creative Writing | 1092 | 1479 |
| LMArena Multi-Turn | 1117 | 1489 |
| EQ-Bench Creative Writing | 451 | — |
Frequently asked questions
Is GPT-3.5-turbo better than Qwen3.8 Max?
Qwen3.8 Max is the stronger model overall, scoring 56.8 to 23.2 on the Noometry Index. GPT-3.5-turbo costs 4.0× less per token, which makes it the better buy when Qwen3.8 Max's lead doesn't matter for your workload.
Which is cheaper, GPT-3.5-turbo or Qwen3.8 Max?
GPT-3.5-turbo is cheaper. It lists at $0.50 per million input tokens and $1.50 per million output tokens; Qwen3.8 Max lists at $2 and $6.
Is GPT-3.5-turbo or Qwen3.8 Max better for coding?
Qwen3.8 Max scores higher on coding benchmarks: 53.5 versus 23.9 in the Noometry coding category.
Which has the bigger context window?
Qwen3.8 Max does, with 1M tokens against 16K.
How many benchmarks do GPT-3.5-turbo and Qwen3.8 Max share?
25 benchmarks have published results for both models. GPT-3.5-turbo has 44 scored results on Noometry and Qwen3.8 Max has 39.