# GPT-5.4 mini vs Qwen3.5-Flash

> GPT-5.4 mini is the stronger model overall, scoring 44.8 to 42.3 on the Noometry Index. Qwen3.5-Flash costs 9.6× less per token, which makes it the better buy when GPT-5.4 mini's lead doesn't matter for your workload.

- Canonical page: https://noometry.com/compare/gpt-5-4-mini-vs-qwen3-5-flash
- Last updated: 2026-10-10
- Shared benchmarks: 31

## Summary

- They share 31 benchmarks with published results for both. GPT-5.4 mini scores higher in 7 categories and Qwen3.5-Flash in 1 category; 7 gaps are clear of the uncertainty.
- The widest gap is in coding, where GPT-5.4 mini leads 45.0 to 33.9.
- The biggest single-benchmark swing is FrontierMath (Tiers 1-3): 51.2% for GPT-5.4 mini and 18.2% for Qwen3.5-Flash.
- Qwen3.5-Flash is cheaper at $0.10 / $0.40 per million input/output tokens, against $0.75 / $4.50 for GPT-5.4 mini.
- Qwen3.5-Flash accepts more context: 1M tokens versus 400K.

## Snapshot

| | GPT-5.4 mini | Qwen3.5-Flash |
|---|---|---|
| Provider | OpenAI | Alibaba (Qwen) |
| Noometry Index | 44.8 | 42.3 |
| Rank | 69 | 105 |
| Context | 400K | 1M |
| Input $/M | $0.75 | $0.10 |
| Output $/M | $4.50 | $0.40 |
| Weights | Proprietary | Proprietary |

## Coding

- GPT-5.4 mini: 45.0 (#67)
- Qwen3.5-Flash: 33.9 (#234)

| Benchmark | GPT-5.4 mini | Qwen3.5-Flash |
|---|---|---|
| LMArena WebDev | 1397 | 1243 |
| LMArena Coding | 1438 | 1412 |
| ALE-Bench | 1,189 | 221.8 |
| FrontierCode | 27% | — |
| SciCode | 49.9% | — |
| WeirdML | 60.3% | — |

## Agentic & Tool Use

- GPT-5.4 mini: 29.6 (#74)
- Qwen3.5-Flash: —

| Benchmark | GPT-5.4 mini | Qwen3.5-Flash |
|---|---|---|
| DeepResearch Bench | 36.3% | — |
| Vending-Bench 2 | — | 462.69 |

## Reasoning

- GPT-5.4 mini: 30.3 (#77)
- Qwen3.5-Flash: 32.9 (#63)

| Benchmark | GPT-5.4 mini | Qwen3.5-Flash |
|---|---|---|
| Chess Puzzles | 24% | 21% |
| LMArena Hard Prompts | 1424 | 1404 |
| Mystery Game Puzzles | 11% | 20% |
| DTBench | 80% | 82.9% |
| LMCA | 40.8% | 29.1% |
| Epoch Capabilities Index | 148.83 | 143.98 |
| ARC-AGI-2 | 18.9% | — |
| Kagi LLM Benchmark | 37.9% | — |
| NYT Connections (extended) | 61.8% | — |
| ARC-AGI-1 | 63.7% | — |
| CritPt | 10% | — |
| Thematic Generalization | 61.7% | — |
| ForecastBench | 57 | — |

## Math

- GPT-5.4 mini: 45.4 (#63)
- Qwen3.5-Flash: 36.3 (#117)

| Benchmark | GPT-5.4 mini | Qwen3.5-Flash |
|---|---|---|
| FrontierMath (Tiers 1-3) | 51.2% | 18.2% |
| OTIS Mock AIME 2024-2025 | 88.9% | 84.4% |
| LMArena Math | 1418 | 1407 |
| FrontierMath (Feb 2025 set) | 28.3% | 6.2% |
| FrontierMath Tier 4 (v1) | 2.1% | 0% |
| FrontierMath Tier 4 | 9.8% | — |
| ProofBench | 21% | — |

## Knowledge

- GPT-5.4 mini: 51.4 (#62)
- Qwen3.5-Flash: 43.2 (#84)

| Benchmark | GPT-5.4 mini | Qwen3.5-Flash |
|---|---|---|
| GPQA Diamond | 86.9% | 82.3% |
| SimpleQA Verified | 29.4% | 20.3% |
| Vectara Hallucination Rate | 5.5% | 10.5% |
| LMArena Expert | 1434 | 1408 |

## Multimodal

- GPT-5.4 mini: 39.0 (#51)
- Qwen3.5-Flash: —

| Benchmark | GPT-5.4 mini | Qwen3.5-Flash |
|---|---|---|
| LMArena Vision | 1245 | — |

## Multilingual

- GPT-5.4 mini: 52.1 (#94)
- Qwen3.5-Flash: 50.7 (#116)

| Benchmark | GPT-5.4 mini | Qwen3.5-Flash |
|---|---|---|
| LMArena Non-English | 1405 | 1385 |
| LMArena Chinese | 1446 | 1446 |
| LMArena French | 1440 | 1412 |
| LMArena German | 1407 | 1389 |
| LMArena Japanese | 1375 | 1369 |
| LMArena Korean | 1369 | 1344 |
| LMArena Russian | 1417 | 1379 |
| LMArena Spanish | 1403 | 1399 |

## Instruction Following

- GPT-5.4 mini: 74.3 (#99)
- Qwen3.5-Flash: 72.8 (#133)

| Benchmark | GPT-5.4 mini | Qwen3.5-Flash |
|---|---|---|
| LMArena Instruction Following | 1405 | 1374 |

## Long Context

- GPT-5.4 mini: 42.9 (#107)
- Qwen3.5-Flash: 42.4 (#119)

| Benchmark | GPT-5.4 mini | Qwen3.5-Flash |
|---|---|---|
| LMArena Longer Query | 1407 | 1392 |

## Writing & Preference

- GPT-5.4 mini: 64.2 (#56)
- Qwen3.5-Flash: 58.2 (#117)

| Benchmark | GPT-5.4 mini | Qwen3.5-Flash |
|---|---|---|
| LMArena Text | 1411 | 1397 |
| LMArena Creative Writing | 1370 | 1343 |
| LMArena Multi-Turn | 1429 | 1393 |
| EQ-Bench Creative Writing | 1665 | — |

## FAQ

### Is GPT-5.4 mini better than Qwen3.5-Flash?

GPT-5.4 mini is the stronger model overall, scoring 44.8 to 42.3 on the Noometry Index. Qwen3.5-Flash costs 9.6× less per token, which makes it the better buy when GPT-5.4 mini's lead doesn't matter for your workload.

### Which is cheaper, GPT-5.4 mini or Qwen3.5-Flash?

Qwen3.5-Flash is cheaper. It lists at $0.10 per million input tokens and $0.40 per million output tokens; GPT-5.4 mini lists at $0.75 and $4.50.

### Is GPT-5.4 mini or Qwen3.5-Flash better for coding?

GPT-5.4 mini scores higher on coding benchmarks: 45.0 versus 33.9 in the Noometry coding category.

### Which has the bigger context window?

Qwen3.5-Flash does, with 1M tokens against 400K.

### How many benchmarks do GPT-5.4 mini and Qwen3.5-Flash share?

31 benchmarks have published results for both models. GPT-5.4 mini has 46 scored results on Noometry and Qwen3.5-Flash has 32.