DeepSeek·released 2024-06-13
Scored on 4 benchmarks. 0% of its scores are independently reproduced — the rest are self-reported or unverified.
Every score, ranked, colored by evidence status.
| Benchmark | Score | Evidence status | Measured |
|---|---|---|---|
| GSM8K | 67.1accuracy (%) | self-reported· optimized | 2024-06-13 |
| MMLU | 47.33accuracy (%) | self-reported· optimized | 2024-06-13 |
| Winogrande | 45.8accuracy (%) | self-reported· optimized | 2024-06-13 |
| ARC AI2 | 43.07accuracy (%) | self-reported· optimized | 2024-06-13 |