Model comparison
GPT-5.4 mini vs Qwen3.6 Max Preview
Qwen3.6 Max Preview is the stronger model overall, scoring 51.1 to 44.8 on the Noometry Index. GPT-5.4 mini costs 1.7× less per token, which makes it the better buy when Qwen3.6 Max Preview's lead doesn't matter for your workload.
Last verified . 26 shared benchmarks.
Alibaba (Qwen)Summary
- They share 26 benchmarks with published results for both. GPT-5.4 mini scores higher in 1 category and Qwen3.6 Max Preview in 7 categories; 7 gaps are clear of the uncertainty.
- The widest gap is in reasoning, where Qwen3.6 Max Preview leads 40.8 to 30.3.
- The biggest single-benchmark swing is SimpleQA Verified: 29.4% for GPT-5.4 mini and 52% for Qwen3.6 Max Preview.
- GPT-5.4 mini is cheaper at $0.75 / $4.50 per million input/output tokens, against $1.30 / $7.80 for Qwen3.6 Max Preview.
- GPT-5.4 mini accepts more context: 400K tokens versus 262K.
Side by side
| GPT-5.4 mini | Qwen3.6 Max Preview | |
|---|---|---|
| Provider | OpenAI | Alibaba (Qwen) |
| Noometry Index | 44.8 | 51.1 |
| Released | 2026-03-17 | 2026-04-20 |
| Weights | Proprietary | Proprietary |
| Context window | 400K | 262K |
| Max output | 128K | 66K |
| Input $ / M tokens | $0.75 | $1.30 |
| Output $ / M tokens | $4.50 | $7.80 |
| Results tracked | 46 | 29 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding Qwen3.6 Max Preview leads
GPT-5.4 mini: 45.0 (#67), Qwen3.6 Max Preview: 48.6 (#50)
| Benchmark | GPT-5.4 mini | Qwen3.6 Max Preview |
|---|---|---|
| LMArena WebDev | 1397 | 1482 |
| LMArena Coding | 1438 | 1472 |
| SWE-bench Verified | — | 76.7% |
| FrontierCode | 27% | — |
| SciCode | 49.9% | — |
| WeirdML | 60.3% | — |
| ALE-Bench | 1,189 | — |
Agentic & Tool Use Not comparable
GPT-5.4 mini: 29.6 (#74), Qwen3.6 Max Preview: —
| Benchmark | GPT-5.4 mini | Qwen3.6 Max Preview |
|---|---|---|
| DeepResearch Bench | 36.3% | — |
| Vending-Bench 2 | — | 4,254 |
Reasoning Qwen3.6 Max Preview leads
GPT-5.4 mini: 30.3 (#77), Qwen3.6 Max Preview: 40.8 (#47)
| Benchmark | GPT-5.4 mini | Qwen3.6 Max Preview |
|---|---|---|
| NYT Connections (extended) | 61.8% | 74.1% |
| Chess Puzzles | 24% | 20% |
| LMArena Hard Prompts | 1424 | 1457 |
| Mystery Game Puzzles | 11% | 19% |
| DTBench | 80% | 87.2% |
| LMCA | 40.8% | 42.5% |
| Epoch Capabilities Index | 148.83 | 149.27 |
| ARC-AGI-2 | 18.9% | — |
| SimpleBench | — | 63% |
| Kagi LLM Benchmark | 37.9% | — |
| ARC-AGI-1 | 63.7% | — |
| CritPt | 10% | — |
| Thematic Generalization | 61.7% | — |
| ForecastBench | 57 | — |
Math Qwen3.6 Max Preview leads
GPT-5.4 mini: 45.4 (#63), Qwen3.6 Max Preview: 52.2 (#40)
| Benchmark | GPT-5.4 mini | Qwen3.6 Max Preview |
|---|---|---|
| OTIS Mock AIME 2024-2025 | 88.9% | 91.1% |
| LMArena Math | 1418 | 1465 |
| FrontierMath (Feb 2025 set) | 28.3% | 23.1% |
| FrontierMath Tier 4 (v1) | 2.1% | 4.2% |
| FrontierMath (Tiers 1-3) | 51.2% | — |
| FrontierMath Tier 4 | 9.8% | — |
| ProofBench | 21% | — |
Knowledge Qwen3.6 Max Preview leads
GPT-5.4 mini: 51.4 (#62), Qwen3.6 Max Preview: 57.6 (#35)
| Benchmark | GPT-5.4 mini | Qwen3.6 Max Preview |
|---|---|---|
| GPQA Diamond | 86.9% | 87.4% |
| SimpleQA Verified | 29.4% | 52% |
| LMArena Expert | 1434 | 1476 |
| Vectara Hallucination Rate | 5.5% | — |
Multimodal Not comparable
GPT-5.4 mini: 39.0 (#51), Qwen3.6 Max Preview: —
| Benchmark | GPT-5.4 mini | Qwen3.6 Max Preview |
|---|---|---|
| LMArena Vision | 1245 | — |
Multilingual Qwen3.6 Max Preview leads
GPT-5.4 mini: 52.1 (#94), Qwen3.6 Max Preview: 54.3 (#46)
| Benchmark | GPT-5.4 mini | Qwen3.6 Max Preview |
|---|---|---|
| LMArena Non-English | 1405 | 1436 |
| LMArena Chinese | 1446 | 1487 |
| LMArena French | 1440 | 1449 |
| LMArena Russian | 1417 | 1444 |
| LMArena Spanish | 1403 | 1452 |
| LMArena German | 1407 | — |
| LMArena Japanese | 1375 | — |
| LMArena Korean | 1369 | — |
Instruction Following Qwen3.6 Max Preview leads
GPT-5.4 mini: 74.3 (#99), Qwen3.6 Max Preview: 75.8 (#51)
| Benchmark | GPT-5.4 mini | Qwen3.6 Max Preview |
|---|---|---|
| LMArena Instruction Following | 1405 | 1437 |
Long Context Qwen3.6 Max Preview leads
GPT-5.4 mini: 42.9 (#107), Qwen3.6 Max Preview: 44.6 (#59)
| Benchmark | GPT-5.4 mini | Qwen3.6 Max Preview |
|---|---|---|
| LMArena Longer Query | 1407 | 1457 |
Writing & Preference Too close to call
GPT-5.4 mini: 64.2 (#56), Qwen3.6 Max Preview: 64.0 (#58)
| Benchmark | GPT-5.4 mini | Qwen3.6 Max Preview |
|---|---|---|
| LMArena Text | 1411 | 1447 |
| LMArena Creative Writing | 1370 | 1435 |
| LMArena Multi-Turn | 1429 | 1456 |
| EQ-Bench Creative Writing | 1665 | — |
Frequently asked questions
Is GPT-5.4 mini better than Qwen3.6 Max Preview?
Qwen3.6 Max Preview is the stronger model overall, scoring 51.1 to 44.8 on the Noometry Index. GPT-5.4 mini costs 1.7× less per token, which makes it the better buy when Qwen3.6 Max Preview's lead doesn't matter for your workload.
Which is cheaper, GPT-5.4 mini or Qwen3.6 Max Preview?
GPT-5.4 mini is cheaper. It lists at $0.75 per million input tokens and $4.50 per million output tokens; Qwen3.6 Max Preview lists at $1.30 and $7.80.
Is GPT-5.4 mini or Qwen3.6 Max Preview better for coding?
Qwen3.6 Max Preview scores higher on coding benchmarks: 48.6 versus 45.0 in the Noometry coding category.
Which has the bigger context window?
GPT-5.4 mini does, with 400K tokens against 262K.
How many benchmarks do GPT-5.4 mini and Qwen3.6 Max Preview share?
26 benchmarks have published results for both models. GPT-5.4 mini has 46 scored results on Noometry and Qwen3.6 Max Preview has 29.