Up to four models
Claude 2.1 vs Claude 2
On the index (GPT-4 = 100), Claude 2.1 scores 96 and Claude 2 94. Side by side on the same third-party tests, with price and context. The link in your address bar shares this exact comparison.
Claude 2.1 and Claude 2 are too close to call (2 points apart, within the uncertainty), ahead on 0 of 1 shared core tests.
Test by testthird-party results
Claude 2.1Claude 2
Every result as a table
| Test | Claude 2.1 | Claude 2 |
|---|---|---|
| Reasoning & knowledge | ||
| GPQA Diamond (Epoch AI) | 33.0% | 34.7% |
| Maths | ||
| AIME-style maths (OTIS mock) | 1.9% | 2.5% |
| Agents | ||
| DTBench | 50.9% | 51.8% |
Tests at least two of these models have taken. Bold rows are the core tests.