# GPT-5 Mini vs Qwen3 Max

> Qwen3 Max is the stronger model overall, scoring 43.7 to 41.8 on the Noometry Index. GPT-5 Mini costs 3.5× less per token, which makes it the better buy when Qwen3 Max's lead doesn't matter for your workload.

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

## Summary

- They share 31 benchmarks with published results for both. GPT-5 Mini scores higher in 4 categories and Qwen3 Max in 4 categories; 7 gaps are clear of the uncertainty.
- The widest gap is in math, where GPT-5 Mini leads 46.7 to 38.7.
- The biggest single-benchmark swing is FrontierMath (Tiers 1-3): 46.7% for GPT-5 Mini and 18.9% for Qwen3 Max.
- GPT-5 Mini is cheaper at $0.25 / $2 per million input/output tokens, against $1.20 / $6 for Qwen3 Max.
- GPT-5 Mini accepts more context: 400K tokens versus 262K.

## Snapshot

| | GPT-5 Mini | Qwen3 Max |
|---|---|---|
| Provider | OpenAI | Alibaba (Qwen) |
| Noometry Index | 41.8 | 43.7 |
| Rank | 128 | 87 |
| Context | 400K | 262K |
| Input $/M | $0.25 | $1.20 |
| Output $/M | $2 | $6 |
| Weights | Proprietary | Proprietary |

## Coding

- GPT-5 Mini: 40.1 (#146)
- Qwen3 Max: 43.0 (#93)

| Benchmark | GPT-5 Mini | Qwen3 Max |
|---|---|---|
| LMArena Coding | 1406 | 1456 |
| ALE-Bench | 799.77 | 370.45 |
| SWE-bench Verified | 64.7% | — |
| SWE-bench Verified (bash only) | 59.8% | — |
| SWE-bench Multilingual | 39.7% | — |
| SciCode | 39.2% | — |
| WeirdML | 52.7% | — |
| AlgoTune | 1.38 | — |

## Agentic & Tool Use

- GPT-5 Mini: 31.1 (#70)
- Qwen3 Max: —

| Benchmark | GPT-5 Mini | Qwen3 Max |
|---|---|---|
| Vending-Bench 2 | -31.18 | 71.56 |
| Terminal-Bench | 34.8% | — |
| Berkeley Function Calling Leaderboard | 55.5% | — |

## Reasoning

- GPT-5 Mini: 23.9 (#168)
- Qwen3 Max: 22.6 (#190)

| Benchmark | GPT-5 Mini | Qwen3 Max |
|---|---|---|
| Kagi LLM Benchmark | 70.3% | 72.5% |
| Chess Puzzles | 30% | 4% |
| LMArena Hard Prompts | 1380 | 1448 |
| Mystery Game Puzzles | 10% | 5% |
| DTBench | 80.5% | 82.1% |
| LMCA | 34.2% | 28.3% |
| Epoch Capabilities Index | 145.52 | 142.38 |
| ARC-AGI-2 | 4.4% | — |
| NYT Connections (extended) | — | 30.1% |
| ARC-AGI-1 | 54.3% | — |
| CritPt | 0% | — |
| EnigmaEval | 8.2% | — |
| ForecastBench | 61 | — |

## Math

- GPT-5 Mini: 46.7 (#69)
- Qwen3 Max: 38.7 (#131)

| Benchmark | GPT-5 Mini | Qwen3 Max |
|---|---|---|
| FrontierMath (Tiers 1-3) | 46.7% | 18.9% |
| OTIS Mock AIME 2024-2025 | 86.7% | 73.3% |
| LMArena Math | 1378 | 1446 |
| MATH Level 5 | 97.8% | 97.1% |
| FrontierMath Tier 4 | 12.2% | — |
| ProofBench | 9% | — |
| Omni-MATH | 72.2% | — |
| FrontierMath (Feb 2025 set) | 27.2% | — |
| FrontierMath Tier 4 (v1) | 6.3% | — |

## Knowledge

- GPT-5 Mini: 45.6 (#86)
- Qwen3 Max: 48.1 (#78)

| Benchmark | GPT-5 Mini | Qwen3 Max |
|---|---|---|
| GPQA Diamond | 75% | 72.6% |
| SimpleQA Verified | 21.6% | 48.7% |
| LMArena Expert | 1379 | 1455 |
| Humanity's Last Exam | 19.4% | — |
| MMLU-Pro | 83.5% | — |
| Confabulations | 13.3% | — |
| Vectara Hallucination Rate | 12.9% | — |
| GPQA (HELM) | 75.6% | — |

## Multimodal

- GPT-5 Mini: 35.6 (#85)
- Qwen3 Max: —

| Benchmark | GPT-5 Mini | Qwen3 Max |
|---|---|---|
| LMArena Vision | 1202 | — |
| VPCT | 40.2% | — |

## Multilingual

- GPT-5 Mini: 48.9 (#137)
- Qwen3 Max: 53.7 (#62)

| Benchmark | GPT-5 Mini | Qwen3 Max |
|---|---|---|
| LMArena Non-English | 1363 | 1429 |
| LMArena Chinese | 1385 | 1478 |
| LMArena French | 1386 | 1449 |
| LMArena German | 1366 | 1463 |
| LMArena Japanese | 1341 | 1397 |
| LMArena Korean | 1308 | 1399 |
| LMArena Russian | 1362 | 1428 |
| LMArena Spanish | 1355 | 1462 |

## Instruction Following

- GPT-5 Mini: 76.2 (#46)
- Qwen3 Max: 74.8 (#87)

| Benchmark | GPT-5 Mini | Qwen3 Max |
|---|---|---|
| LMArena Instruction Following | 1357 | 1419 |
| IFEval | 92.7% | — |

## Long Context

- GPT-5 Mini: 41.9 (#132)
- Qwen3 Max: 41.6 (#134)

| Benchmark | GPT-5 Mini | Qwen3 Max |
|---|---|---|
| Fiction.LiveBench | 69.4% | 66.7% |
| LMArena Longer Query | 1355 | 1438 |
| CL-bench | — | 14.5% |

## Writing & Preference

- GPT-5 Mini: 55.2 (#148)
- Qwen3 Max: 62.4 (#76)

| Benchmark | GPT-5 Mini | Qwen3 Max |
|---|---|---|
| LMArena Text | 1373 | 1439 |
| LMArena Creative Writing | 1325 | 1402 |
| LMArena Multi-Turn | 1363 | 1446 |
| Short-Story Creative Writing | 83.1% | — |
| EQ-Bench Creative Writing | 1313 | — |
| WildBench | 85.5% | — |

## FAQ

### Is GPT-5 Mini better than Qwen3 Max?

Qwen3 Max is the stronger model overall, scoring 43.7 to 41.8 on the Noometry Index. GPT-5 Mini costs 3.5× less per token, which makes it the better buy when Qwen3 Max's lead doesn't matter for your workload.

### Which is cheaper, GPT-5 Mini or Qwen3 Max?

GPT-5 Mini is cheaper. It lists at $0.25 per million input tokens and $2 per million output tokens; Qwen3 Max lists at $1.20 and $6.

### Is GPT-5 Mini or Qwen3 Max better for coding?

Qwen3 Max scores higher on coding benchmarks: 43.0 versus 40.1 in the Noometry coding category.

### Which has the bigger context window?

GPT-5 Mini does, with 400K tokens against 262K.

### How many benchmarks do GPT-5 Mini and Qwen3 Max share?

31 benchmarks have published results for both models. GPT-5 Mini has 60 scored results on Noometry and Qwen3 Max has 33.