Up to four models
Llama 2-13B vs Llama 3.2 11B
On the index (GPT-4 = 100), Llama 2-13B scores 87 and Llama 3.2 11B 83. Side by side on the same third-party tests, with price and context. The link in your address bar shares this exact comparison.
Llama 2-13B and Llama 3.2 11B are too close to call (4 points apart, within the uncertainty).
Test by testthird-party results
Every result as a table
| Test | Llama 2-13B | Llama 3.2 11B |
|---|---|---|
| Reasoning & knowledge | ||
| MMLU (HELM runs) | 55.4% | 56.5% |
Tests at least two of these models have taken. Bold rows are the core tests.