Up to four models
o4-mini vs Qwen3.6 27B
On the index (GPT-4 = 100), o4-mini scores 141 and Qwen3.6 27B 141. Side by side on the same third-party tests, with price and context. The link in your address bar shares this exact comparison.
141125–158
WeightsOpen
ReleasedApr 2026
Price in / out$0.32 / $3.2
Context262K
Core tests taken2 of 9
o4-mini and Qwen3.6 27B are too close to call (0 points apart, within the uncertainty), ahead on 1 of 2 shared core tests.
Test by testthird-party results
o4-mini
Qwen3.6 27B
Every result as a table
| Test | o4-mini | Qwen3.6 27B |
|---|---|---|
| Reasoning & knowledge | ||
| GPQA Diamond (Epoch AI) | 79.6% | 85.9% |
| Maths | ||
| FrontierMath (tiers 1-3) | 36.1% | 35.1% |
| AIME-style maths (OTIS mock) | 81.7% | 91.1% |
| Agents | ||
| DTBench | 77.6% | 78.1% |
| LMCA | 26.5% | 34.5% |
| Mystery game puzzles | 5.0% | 7.0% |
| Professional work | ||
| CorpFin v2 | 59.0% | 62.3% |
| MortgageTax | 64.8% | 68.3% |
| TaxEval v2 | 74.8% | 71.3% |
Task length · METR38 sec
Tests at least two of these models have taken. Bold rows are the core tests.