janus-dpo-7B: benchmark results
Provider: KAIST. Released 2024-04-25. Access: Open.
Unified ELO 1408 ± 1, rank #1104 of 1392 rated models, from 14 benchmark results.
Strongest benchmark results
| Benchmark | Score | Metric | Percentile |
|---|---|---|---|
| Open LLM Leaderboard - MuSR | 13.71 | Score | 75.4 |
| Open Chinese LLM - TruthfulQA MC | 52.92 | Accuracy (%) | 43.9 |
| Open LLM Leaderboard - BBH | 27.09 | Score | 42.9 |
| Open LLM Leaderboard - IFEval | 40.03 | Score | 39.9 |
| Open LLM Leaderboard - GPQA | 4.25 | Score | 36.8 |
| Open LLM Leaderboard - MMLU-Pro | 21.96 | Score | 35.6 |
| Open Chinese LLM - GSM8K | 31.77 | Accuracy (%) | 26.4 |
| Open Chinese LLM - CMMLU | 47.35 | Accuracy (%) | 25.5 |
| Open Chinese LLM - HellaSwag | 55.16 | Accuracy (%) | 25.2 |
| Open Chinese LLM - ARC Challenge | 46.67 | Accuracy (%) | 24 |
| Open LLM Leaderboard - MATH Level 5 | 4.15 | Score | 23.9 |
| Open Chinese LLM - WinoGrande | 61.01 | Accuracy (%) | 21.8 |
Interactive version: theaggregate.ai/model?slug=janus-dpo-7b · How It Works · Data refreshed daily, snapshot 2026-09-05.