# o3-mini vs Qwen2.5-Max

> Qwen2.5-Max is the stronger model overall, scoring 40.1 to 36.7 on the Noometry Index.

- Canonical page: https://noometry.com/compare/o3-mini-vs-qwen2-5-max
- Last updated: 2026-10-10
- Shared benchmarks: 27

## Summary

- They share 27 benchmarks with published results for both. o3-mini scores higher in 3 categories and Qwen2.5-Max in 5 categories; 7 gaps are clear of the uncertainty.
- The widest gap is in reasoning, where Qwen2.5-Max leads 24.6 to 16.3.
- The biggest single-benchmark swing is LiveBench Reasoning: 89.6% for o3-mini and 51.4% for Qwen2.5-Max.

## Snapshot

| | o3-mini | Qwen2.5-Max |
|---|---|---|
| Provider | OpenAI | Alibaba (Qwen) |
| Noometry Index | 36.7 | 40.1 |
| Rank | 205 | 142 |
| Context | 200K | — |
| Input $/M | $1.10 | — |
| Output $/M | $4.40 | — |
| Weights | Proprietary | Proprietary |

## Coding

- o3-mini: 41.4 (#116)
- Qwen2.5-Max: 41.2 (#118)

| Benchmark | o3-mini | Qwen2.5-Max |
|---|---|---|
| LiveBench Coding | 82.7% | 64.4% |
| LMArena Coding | 1378 | 1359 |
| Aider Polyglot | 60.4% | — |
| SciCode | 39.8% | — |
| GSO | 1.3% | — |
| WeirdML | 43.7% | — |
| CadEval | 54% | — |

## Agentic & Tool Use

- o3-mini: 29.3 (#79)
- Qwen2.5-Max: —

| Benchmark | o3-mini | Qwen2.5-Max |
|---|---|---|
| Cybench | 22.5% | — |

## Reasoning

- o3-mini: 16.3 (#295)
- Qwen2.5-Max: 24.6 (#144)

| Benchmark | o3-mini | Qwen2.5-Max |
|---|---|---|
| LiveBench Reasoning | 89.6% | 51.4% |
| LMArena Hard Prompts | 1366 | 1360 |
| LiveBench Data Analysis | 70.6% | 67.9% |
| Epoch Capabilities Index | 140.35 | 132.54 |
| LiveBench | 75.9% | 62.3% |
| ARC-AGI-2 | 3% | — |
| SimpleBench | 22.8% | — |
| ARC-AGI-1 | 34.5% | — |
| CritPt | 0.3% | — |
| Chess Puzzles | 17% | — |
| Mystery Game Puzzles | 7% | — |
| DTBench | 68.8% | — |
| LMCA | 19% | — |
| ForecastBench | 59.6 | — |

## Math

- o3-mini: 27.6 (#236)
- Qwen2.5-Max: 34.3 (#162)

| Benchmark | o3-mini | Qwen2.5-Max |
|---|---|---|
| LiveBench Math | 77.3% | 58.4% |
| LMArena Math | 1396 | 1369 |
| FrontierMath (Tiers 1-3) | 18.6% | — |
| FrontierMath Tier 4 | 0% | — |
| OTIS Mock AIME 2024-2025 | 76.9% | — |
| MATH Level 5 | 96.5% | — |
| FrontierMath (Feb 2025 set) | 12.4% | — |
| FrontierMath Tier 4 (v1) | 4.2% | — |

## Knowledge

- o3-mini: 38.3 (#135)
- Qwen2.5-Max: 34.8 (#181)

| Benchmark | o3-mini | Qwen2.5-Max |
|---|---|---|
| Confabulations | 17.9% | 21.8% |
| LMArena Expert | 1364 | 1337 |
| GPQA Diamond | 77% | — |
| SimpleQA Verified | 15.3% | — |

## Multilingual

- o3-mini: 45.9 (#159)
- Qwen2.5-Max: 48.3 (#140)

| Benchmark | o3-mini | Qwen2.5-Max |
|---|---|---|
| LMArena Non-English | 1319 | 1352 |
| LMArena Chinese | 1379 | 1382 |
| LMArena French | 1333 | 1396 |
| LMArena German | 1303 | 1350 |
| LMArena Japanese | 1286 | 1299 |
| LMArena Korean | 1314 | 1304 |
| LMArena Russian | 1304 | 1353 |
| LMArena Spanish | 1321 | 1377 |

## Instruction Following

- o3-mini: 75.3 (#67)
- Qwen2.5-Max: 71.5 (#147)

| Benchmark | o3-mini | Qwen2.5-Max |
|---|---|---|
| LiveBench Instruction Following | 84.4% | 75.3% |
| LMArena Instruction Following | 1338 | 1335 |

## Long Context

- o3-mini: 33.8 (#251)
- Qwen2.5-Max: 41.3 (#137)

| Benchmark | o3-mini | Qwen2.5-Max |
|---|---|---|
| LMArena Longer Query | 1343 | 1358 |
| Fiction.LiveBench | 50% | — |

## Writing & Preference

- o3-mini: 50.4 (#177)
- Qwen2.5-Max: 55.6 (#141)

| Benchmark | o3-mini | Qwen2.5-Max |
|---|---|---|
| LMArena Text | 1337 | 1367 |
| LMArena Creative Writing | 1286 | 1339 |
| Short-Story Creative Writing | 61.7% | 72.9% |
| LMArena Multi-Turn | 1320 | 1364 |
| LiveBench Language | 50.7% | 56.3% |

## FAQ

### Is o3-mini better than Qwen2.5-Max?

Qwen2.5-Max is the stronger model overall, scoring 40.1 to 36.7 on the Noometry Index.

### Is o3-mini or Qwen2.5-Max better for coding?

They score almost the same on coding (41.4 vs 41.2); test both on your own repository before choosing.

### How many benchmarks do o3-mini and Qwen2.5-Max share?

27 benchmarks have published results for both models. o3-mini has 51 scored results on Noometry and Qwen2.5-Max has 27.