# GPT-4.1 mini vs Qwen2-72B

> GPT-4.1 mini is the stronger model overall, scoring 33.5 to 29.5 on the Noometry Index.

- Canonical page: https://noometry.com/compare/gpt-4-1-mini-vs-qwen2-72b
- Last updated: 2026-10-10
- Shared benchmarks: 22

## Summary

- They share 22 benchmarks with published results for both. GPT-4.1 mini scores higher in 6 categories and Qwen2-72B in 3 categories; 9 gaps are clear of the uncertainty.
- The widest gap is in agentic & tool use, where GPT-4.1 mini leads 33.0 to 16.9.
- The biggest single-benchmark swing is MATH Level 5: 87.3% for GPT-4.1 mini and 39.1% for Qwen2-72B.
- Qwen2-72B has downloadable open weights; the other is API-only.

## Snapshot

| | GPT-4.1 mini | Qwen2-72B |
|---|---|---|
| Provider | OpenAI | Alibaba (Qwen) |
| Noometry Index | 33.5 | 29.5 |
| Rank | 232 | 294 |
| Context | 1.05M | — |
| Input $/M | $0.40 | — |
| Output $/M | $1.60 | — |
| Weights | Proprietary | Open |

## Coding

- GPT-4.1 mini: 30.4 (#286)
- Qwen2-72B: 28.7 (#303)

| Benchmark | GPT-4.1 mini | Qwen2-72B |
|---|---|---|
| WeirdML | 37.6% | 11.3% |
| BigCodeBench Instruct | 48.9% | 38.5% |
| LMArena Coding | 1367 | 1196 |
| SWE-bench Verified (bash only) | 23.9% | — |
| Aider Polyglot | 32.4% | — |
| SciCode | 40.4% | — |
| BigCodeBench Complete | — | 54% |
| CadEval | 16% | — |

## Agentic & Tool Use

- GPT-4.1 mini: 33.0 (#52)
- Qwen2-72B: 16.9 (#138)

| Benchmark | GPT-4.1 mini | Qwen2-72B |
|---|---|---|
| Berkeley Function Calling Leaderboard | 50.5% | — |
| TheAgentCompany | — | 1.1% |
| METR Time Horizons | — | 29.9% |

## Reasoning

- GPT-4.1 mini: 10.9 (#335)
- Qwen2-72B: 22.2 (#184)

| Benchmark | GPT-4.1 mini | Qwen2-72B |
|---|---|---|
| LMArena Hard Prompts | 1349 | 1191 |
| Epoch Capabilities Index | 135.01 | 125.27 |
| ARC-AGI-2 | 0% | — |
| Kagi LLM Benchmark | 48.6% | — |
| ARC-AGI-1 | 3.5% | — |
| CritPt | 0% | — |
| Chess Puzzles | 7% | — |
| Mystery Game Puzzles | 7% | — |
| DTBench | 68.8% | — |
| LMCA | 21.1% | — |

## Math

- GPT-4.1 mini: 23.1 (#264)
- Qwen2-72B: 27.9 (#232)

| Benchmark | GPT-4.1 mini | Qwen2-72B |
|---|---|---|
| LMArena Math | 1343 | 1235 |
| MATH Level 5 | 87.3% | 39.1% |
| FrontierMath (Tiers 1-3) | 6.7% | — |
| OTIS Mock AIME 2024-2025 | 44.7% | — |
| Omni-MATH | 49.1% | — |
| FrontierMath (Feb 2025 set) | 4.5% | — |

## Knowledge

- GPT-4.1 mini: 34.7 (#183)
- Qwen2-72B: 21.2 (#270)

| Benchmark | GPT-4.1 mini | Qwen2-72B |
|---|---|---|
| GPQA Diamond | 65.8% | 40.8% |
| LMArena Expert | 1338 | 1171 |
| SimpleQA Verified | 12.7% | — |
| MMLU-Pro | 78.3% | — |
| GPQA (HELM) | 61.4% | — |
| MMLU | — | 82.4% |

## Multimodal

- GPT-4.1 mini: 35.2 (#77)
- Qwen2-72B: —

| Benchmark | GPT-4.1 mini | Qwen2-72B |
|---|---|---|
| LMArena Vision | 1181 | — |

## Multilingual

- GPT-4.1 mini: 45.9 (#161)
- Qwen2-72B: 36.0 (#239)

| Benchmark | GPT-4.1 mini | Qwen2-72B |
|---|---|---|
| LMArena Non-English | 1318 | 1176 |
| LMArena Chinese | 1329 | 1240 |
| LMArena French | 1357 | 1169 |
| LMArena German | 1349 | 1151 |
| LMArena Japanese | 1291 | 1111 |
| LMArena Korean | 1298 | 1083 |
| LMArena Russian | 1324 | 1169 |
| LMArena Spanish | 1319 | 1168 |

## Instruction Following

- GPT-4.1 mini: 73.9 (#113)
- Qwen2-72B: 62.0 (#236)

| Benchmark | GPT-4.1 mini | Qwen2-72B |
|---|---|---|
| LMArena Instruction Following | 1332 | 1181 |
| IFEval | 90.4% | — |

## Long Context

- GPT-4.1 mini: 31.8 (#270)
- Qwen2-72B: 36.1 (#230)

| Benchmark | GPT-4.1 mini | Qwen2-72B |
|---|---|---|
| LMArena Longer Query | 1344 | 1193 |
| Fiction.LiveBench | 44.4% | — |

## Writing & Preference

- GPT-4.1 mini: 48.8 (#193)
- Qwen2-72B: 41.0 (#236)

| Benchmark | GPT-4.1 mini | Qwen2-72B |
|---|---|---|
| LMArena Text | 1341 | 1203 |
| LMArena Creative Writing | 1300 | 1181 |
| LMArena Multi-Turn | 1354 | 1195 |
| EQ-Bench Creative Writing | 1147 | — |
| WildBench | 83.8% | — |

## FAQ

### Is GPT-4.1 mini better than Qwen2-72B?

GPT-4.1 mini is the stronger model overall, scoring 33.5 to 29.5 on the Noometry Index.

### Is GPT-4.1 mini or Qwen2-72B better for coding?

GPT-4.1 mini scores higher on coding benchmarks: 30.4 versus 28.7 in the Noometry coding category.

### How many benchmarks do GPT-4.1 mini and Qwen2-72B share?

22 benchmarks have published results for both models. GPT-4.1 mini has 47 scored results on Noometry and Qwen2-72B has 26.