# GPT-5.4 mini vs Qwen3.7 Max

> Qwen3.7 Max is the stronger model overall, scoring 51.4 to 44.8 on the Noometry Index. GPT-5.4 mini costs 2.2× less per token, which makes it the better buy when Qwen3.7 Max's lead doesn't matter for your workload.

- Canonical page: https://noometry.com/compare/gpt-5-4-mini-vs-qwen3-7-max
- Last updated: 2026-10-10
- Shared benchmarks: 28

## Summary

- They share 28 benchmarks with published results for both. GPT-5.4 mini scores higher in 1 category and Qwen3.7 Max in 8 categories; 8 gaps are clear of the uncertainty.
- The widest gap is in reasoning, where Qwen3.7 Max leads 48.4 to 30.3.
- The biggest single-benchmark swing is SimpleQA Verified: 29.4% for GPT-5.4 mini and 55.8% for Qwen3.7 Max.
- GPT-5.4 mini is cheaper at $0.75 / $4.50 per million input/output tokens, against $2.50 / $7.50 for Qwen3.7 Max.
- Qwen3.7 Max accepts more context: 1M tokens versus 400K.

## Snapshot

| | GPT-5.4 mini | Qwen3.7 Max |
|---|---|---|
| Provider | OpenAI | Alibaba (Qwen) |
| Noometry Index | 44.8 | 51.4 |
| Rank | 69 | 35 |
| Context | 400K | 1M |
| Input $/M | $0.75 | $2.50 |
| Output $/M | $4.50 | $7.50 |
| Weights | Proprietary | Proprietary |

## Coding

- GPT-5.4 mini: 45.0 (#67)
- Qwen3.7 Max: 50.2 (#43)

| Benchmark | GPT-5.4 mini | Qwen3.7 Max |
|---|---|---|
| LMArena WebDev | 1397 | 1515 |
| SciCode | 49.9% | 48.8% |
| LMArena Coding | 1438 | 1498 |
| ALE-Bench | 1,189 | 1,189 |
| SWE-bench Verified | — | 77.3% |
| FrontierCode | 27% | — |
| WeirdML | 60.3% | — |

## Agentic & Tool Use

- GPT-5.4 mini: 29.6 (#74)
- Qwen3.7 Max: 21.8 (#129)

| Benchmark | GPT-5.4 mini | Qwen3.7 Max |
|---|---|---|
| DeepResearch Bench | 36.3% | — |
| GBAEval | — | 0.4% |

## Reasoning

- GPT-5.4 mini: 30.3 (#77)
- Qwen3.7 Max: 48.4 (#36)

| Benchmark | GPT-5.4 mini | Qwen3.7 Max |
|---|---|---|
| NYT Connections (extended) | 61.8% | 85.1% |
| CritPt | 10% | 13.4% |
| Chess Puzzles | 24% | 19% |
| LMArena Hard Prompts | 1424 | 1483 |
| Mystery Game Puzzles | 11% | 32% |
| DTBench | 80% | 92.3% |
| LMCA | 40.8% | 44% |
| Epoch Capabilities Index | 148.83 | 153.74 |
| ARC-AGI-2 | 18.9% | — |
| SimpleBench | — | 70.4% |
| Kagi LLM Benchmark | 37.9% | — |
| ARC-AGI-1 | 63.7% | — |
| Thematic Generalization | 61.7% | — |
| EBR-Bench | — | 9.5% |
| ForecastBench | 57 | — |

## Math

- GPT-5.4 mini: 45.4 (#63)
- Qwen3.7 Max: 62.3 (#26)

| Benchmark | GPT-5.4 mini | Qwen3.7 Max |
|---|---|---|
| FrontierMath (Tiers 1-3) | 51.2% | 64.6% |
| FrontierMath Tier 4 | 9.8% | 34.1% |
| OTIS Mock AIME 2024-2025 | 88.9% | 95.6% |
| ProofBench | 21% | 26% |
| LMArena Math | 1418 | 1488 |
| FrontierMath (Feb 2025 set) | 28.3% | — |
| FrontierMath Tier 4 (v1) | 2.1% | — |

## Knowledge

- GPT-5.4 mini: 51.4 (#62)
- Qwen3.7 Max: 61.7 (#24)

| Benchmark | GPT-5.4 mini | Qwen3.7 Max |
|---|---|---|
| GPQA Diamond | 86.9% | 90.9% |
| SimpleQA Verified | 29.4% | 55.8% |
| LMArena Expert | 1434 | 1488 |
| Vectara Hallucination Rate | 5.5% | — |

## Multimodal

- GPT-5.4 mini: 39.0 (#51)
- Qwen3.7 Max: —

| Benchmark | GPT-5.4 mini | Qwen3.7 Max |
|---|---|---|
| LMArena Vision | 1245 | — |

## Multilingual

- GPT-5.4 mini: 52.1 (#94)
- Qwen3.7 Max: 57.1 (#13)

| Benchmark | GPT-5.4 mini | Qwen3.7 Max |
|---|---|---|
| LMArena Non-English | 1405 | 1475 |
| LMArena Chinese | 1446 | 1533 |
| LMArena Russian | 1417 | 1482 |
| LMArena French | 1440 | — |
| LMArena German | 1407 | — |
| LMArena Japanese | 1375 | — |
| LMArena Korean | 1369 | — |
| LMArena Spanish | 1403 | — |

## Instruction Following

- GPT-5.4 mini: 74.3 (#99)
- Qwen3.7 Max: 76.9 (#34)

| Benchmark | GPT-5.4 mini | Qwen3.7 Max |
|---|---|---|
| LMArena Instruction Following | 1405 | 1460 |

## Long Context

- GPT-5.4 mini: 42.9 (#107)
- Qwen3.7 Max: 45.4 (#34)

| Benchmark | GPT-5.4 mini | Qwen3.7 Max |
|---|---|---|
| LMArena Longer Query | 1407 | 1484 |

## Writing & Preference

- GPT-5.4 mini: 64.2 (#56)
- Qwen3.7 Max: 65.2 (#52)

| Benchmark | GPT-5.4 mini | Qwen3.7 Max |
|---|---|---|
| LMArena Text | 1411 | 1476 |
| LMArena Creative Writing | 1370 | 1450 |
| LMArena Multi-Turn | 1429 | 1482 |
| EQ-Bench Creative Writing | 1665 | — |
| EQ-Bench 4 | — | 1110 |

## FAQ

### Is GPT-5.4 mini better than Qwen3.7 Max?

Qwen3.7 Max is the stronger model overall, scoring 51.4 to 44.8 on the Noometry Index. GPT-5.4 mini costs 2.2× less per token, which makes it the better buy when Qwen3.7 Max's lead doesn't matter for your workload.

### Which is cheaper, GPT-5.4 mini or Qwen3.7 Max?

GPT-5.4 mini is cheaper. It lists at $0.75 per million input tokens and $4.50 per million output tokens; Qwen3.7 Max lists at $2.50 and $7.50.

### Is GPT-5.4 mini or Qwen3.7 Max better for coding?

Qwen3.7 Max scores higher on coding benchmarks: 50.2 versus 45.0 in the Noometry coding category.

### Which has the bigger context window?

Qwen3.7 Max does, with 1M tokens against 400K.

### How many benchmarks do GPT-5.4 mini and Qwen3.7 Max share?

28 benchmarks have published results for both models. GPT-5.4 mini has 46 scored results on Noometry and Qwen3.7 Max has 33.