Model comparison
GPT-4 vs Qwen3.6 Max Preview
Qwen3.6 Max Preview is the stronger model overall, scoring 51.5 to 29.1 on the Noometry Index.
Last verified . 21 shared benchmarks.
Alibaba (Qwen)Summary
- They share 21 benchmarks with published results for both. GPT-4 scores higher in 0 categories and Qwen3.6 Max Preview in 8 categories; 8 gaps are clear of the uncertainty.
- The widest gap is in math, where Qwen3.6 Max Preview leads 54.1 to 10.8.
- The biggest single-benchmark swing is OTIS Mock AIME 2024-2025: 1.1% for GPT-4 and 91.1% for Qwen3.6 Max Preview.
- Qwen3.6 Max Preview is cheaper at $1.30 / $7.80 per million input/output tokens, against $30 / $60 for GPT-4.
- Qwen3.6 Max Preview accepts more context: 262K tokens versus 8K.
Side by side
| GPT-4 | Qwen3.6 Max Preview | |
|---|---|---|
| Provider | OpenAI | Alibaba (Qwen) |
| Noometry Index | 29.1 | 51.5 |
| Released | 2023-03-14 | 2026-04-20 |
| Weights | Proprietary | Proprietary |
| Context window | 8K | 262K |
| Max output | 8K | 66K |
| Input $ / M tokens | $30 | $1.30 |
| Output $ / M tokens | $60 | $7.80 |
| Results tracked | 38 | 29 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding Qwen3.6 Max Preview leads
GPT-4: 31.6 (#283), Qwen3.6 Max Preview: 48.7 (#54)
| Benchmark | GPT-4 | Qwen3.6 Max Preview |
|---|---|---|
| LMArena Coding | 1254 | 1471 |
| SWE-bench Verified | — | 76.7% |
| LMArena WebDev | — | 1482 |
| WeirdML | 12.4% | — |
| BigCodeBench Instruct | 46% | — |
| BigCodeBench Complete | 57.2% | — |
| HumanEval+ | 79.3% | — |
Agentic & Tool Use Not comparable
GPT-4: —, Qwen3.6 Max Preview: —
| Benchmark | GPT-4 | Qwen3.6 Max Preview |
|---|---|---|
| METR Time Horizons | 36.1% | — |
| Vending-Bench 2 | — | 4,254 |
Reasoning Qwen3.6 Max Preview leads
GPT-4: 17.8 (#289), Qwen3.6 Max Preview: 41.7 (#53)
| Benchmark | GPT-4 | Qwen3.6 Max Preview |
|---|---|---|
| Chess Puzzles | 4% | 20% |
| LMArena Hard Prompts | 1241 | 1457 |
| Mystery Game Puzzles | 12% | 19% |
| DTBench | 62.7% | 87.2% |
| LMCA | 17.1% | 42.5% |
| Epoch Capabilities Index | 125.89 | 149.24 |
| SimpleBench | — | 63% |
| NYT Connections (extended) | — | 74.1% |
| BIG-Bench Hard | 75.1% | — |
| ForecastBench | 57.8 | — |
| HellaSwag | 95.3% | — |
| WinoGrande | 87.5% | — |
Math Qwen3.6 Max Preview leads
GPT-4: 10.8 (#309), Qwen3.6 Max Preview: 54.1 (#46)
| Benchmark | GPT-4 | Qwen3.6 Max Preview |
|---|---|---|
| OTIS Mock AIME 2024-2025 | 1.1% | 91.1% |
| LMArena Math | 1269 | 1465 |
| MATH Level 5 | 23% | — |
| FrontierMath (Feb 2025 set) | — | 23.1% |
| FrontierMath Tier 4 (v1) | — | 4.2% |
| GSM8K | 92% | — |
Knowledge Qwen3.6 Max Preview leads
GPT-4: 18.4 (#282), Qwen3.6 Max Preview: 57.6 (#39)
| Benchmark | GPT-4 | Qwen3.6 Max Preview |
|---|---|---|
| GPQA Diamond | 35.7% | 87.4% |
| LMArena Expert | 1211 | 1478 |
| SimpleQA Verified | — | 52% |
| MMLU | 86.4% | — |
| TriviaQA | 84.8% | — |
Multilingual Qwen3.6 Max Preview leads
GPT-4: 40.6 (#215), Qwen3.6 Max Preview: 54.2 (#48)
| Benchmark | GPT-4 | Qwen3.6 Max Preview |
|---|---|---|
| LMArena Non-English | 1246 | 1437 |
| LMArena Chinese | 1242 | 1487 |
| LMArena French | 1283 | 1449 |
| LMArena Russian | 1251 | 1445 |
| LMArena Spanish | 1261 | 1454 |
| LMArena German | 1251 | — |
| LMArena Japanese | 1209 | — |
| LMArena Korean | 1184 | — |
Instruction Following Qwen3.6 Max Preview leads
GPT-4: 65.3 (#222), Qwen3.6 Max Preview: 75.7 (#55)
| Benchmark | GPT-4 | Qwen3.6 Max Preview |
|---|---|---|
| LMArena Instruction Following | 1241 | 1438 |
Long Context Qwen3.6 Max Preview leads
GPT-4: 37.7 (#212), Qwen3.6 Max Preview: 44.6 (#61)
| Benchmark | GPT-4 | Qwen3.6 Max Preview |
|---|---|---|
| LMArena Longer Query | 1244 | 1457 |
Writing & Preference Qwen3.6 Max Preview leads
GPT-4: 34.9 (#268), Qwen3.6 Max Preview: 63.8 (#60)
| Benchmark | GPT-4 | Qwen3.6 Max Preview |
|---|---|---|
| LMArena Text | 1263 | 1447 |
| LMArena Creative Writing | 1244 | 1435 |
| LMArena Multi-Turn | 1257 | 1456 |
| EQ-Bench Creative Writing | 752 | — |
Frequently asked questions
Is GPT-4 better than Qwen3.6 Max Preview?
Qwen3.6 Max Preview is the stronger model overall, scoring 51.5 to 29.1 on the Noometry Index.
Which is cheaper, GPT-4 or Qwen3.6 Max Preview?
Qwen3.6 Max Preview is cheaper. It lists at $1.30 per million input tokens and $7.80 per million output tokens; GPT-4 lists at $30 and $60.
Is GPT-4 or Qwen3.6 Max Preview better for coding?
Qwen3.6 Max Preview scores higher on coding benchmarks: 48.7 versus 31.6 in the Noometry coding category.
Which has the bigger context window?
Qwen3.6 Max Preview does, with 262K tokens against 8K.
How many benchmarks do GPT-4 and Qwen3.6 Max Preview share?
21 benchmarks have published results for both models. GPT-4 has 38 scored results on Noometry and Qwen3.6 Max Preview has 29.