Up to four models
o1-mini vs Llama 3.2
On the index (GPT-4 = 100), o1-mini scores 123 and Llama 3.2 106. Side by side on the same third-party tests, with price and context. The link in your address bar shares this exact comparison.
o1-mini leads Llama 3.2 by 17 points, ahead on 1 of 1 shared core tests.
Test by testthird-party results
o1-mini
Llama 3.2
Every result as a table
| Test | o1-mini | Llama 3.2 |
|---|---|---|
| Reasoning & knowledge | ||
| GPQA Diamond (Epoch AI) | 62.4% | 41.0% |
| Maths | ||
| AIME-style maths (OTIS mock) | 46.9% | 2.6% |
| MATH (level 5) | 89.2% | 39.4% |
Tests at least two of these models have taken. Bold rows are the core tests.