Up to four models
DeepSeek-R1-0528 vs Claude Sonnet 4
On the index (GPT-4 = 100), DeepSeek-R1-0528 scores 135 and Claude Sonnet 4 133. Side by side on the same third-party tests, with price and context. The link in your address bar shares this exact comparison.
DeepSeek-R1-0528 and Claude Sonnet 4 are too close to call (2 points apart, within the uncertainty), ahead on 0 of 2 shared core tests.
Test by testthird-party results
Every result as a table
| Test | DeepSeek-R1-0528 | Claude Sonnet 4 |
|---|---|---|
| Reasoning & knowledge | ||
| GPQA Diamond (Epoch AI) | 76.3% | 79.2% |
| Fiction.LiveBench | 75.0% | 46.9% |
| SimpleBench | 40.8% | 45.5% |
| Maths | ||
| AIME-style maths (OTIS mock) | 66.4% | 71.1% |
| MATH (level 5) | 96.6% | 84.4% |
| Coding | ||
| Aider Polyglot | 71.4% | 61.3% |
| WeirdML | 41.6% | 46.1% |
| Agents | ||
| DeepResearch Bench | 35.1% | 46.6% |
Task length · METR
32 sec37 sec
Tests at least two of these models have taken. Bold rows are the core tests.