Up to four models
Claude 2.1 vs Qwen2.5-7B
On the index (GPT-4 = 100), Claude 2.1 scores 96 and Qwen2.5-7B 96. 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 Qwen2.5-7B are too close to call (1 point apart, within the uncertainty), ahead on 0 of 1 shared core tests.
Test by testthird-party results
Claude 2.1
Qwen2.5-7B
Every result as a table
| Test | Claude 2.1 | Qwen2.5-7B |
|---|---|---|
| Reasoning & knowledge | ||
| GPQA Diamond (Epoch AI) | 33.0% | 35.5% |
| MMLU (HELM runs) | 73.5% | 72.9% |
| Maths | ||
| AIME-style maths (OTIS mock) | 1.9% | 2.5% |
| Agents | ||
| DTBench | 50.9% | 47.7% |
Tests at least two of these models have taken. Bold rows are the core tests.