Qwen·released 2023-11-30
Scored on 6 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 |
|---|---|---|---|
| LAMBADA | 58.4accuracy (%) | self-reported· optimized | 2023-11-30 |
| PIQA | 46.6accuracy (%) | self-reported· optimized | 2023-11-30 |
| ARC AI2 | 37.6accuracy (%) | self-reported· optimized | 2023-11-30 |
| GSM8K | 21.2accuracy (%) | self-reported· optimized | 2023-11-30 |
| MMLU | 4.27accuracy (%) | self-reported· optimized | 2023-11-30 |
| BBH | 4.27accuracy (%) | self-reported· optimized | 2023-11-30 |