Up to four models
Llama 2 vs Claude Instant
On the index (GPT-4 = 100), Llama 2 scores 95 and Claude Instant 93. 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 and Claude Instant are too close to call (2 points apart, within the uncertainty).
Test by testthird-party results
Every result as a table
| Test | Llama 2 | Claude Instant |
|---|---|---|
| Reasoning & knowledge | ||
| MMLU (HELM runs) | 69.5% | 68.8% |
| Agents | ||
| DTBench | 41.6% | 45.8% |
Tests at least two of these models have taken. Bold rows are the core tests.