Llama-3-Base-8B-SFT-ORPO: benchmark results
Princeton NLP's ORPO baseline on its Llama 3 8B SFT base, released with the SimPO paper's baseline suite. Provider: Meta. Released 2024-05-17. Access: Open.
Unified ELO 1491 ± 1, rank #822 of 1553 rated models, from 7 benchmark results.
Strongest benchmark results
| Benchmark | Score | Metric | Percentile |
|---|---|---|---|
| Open Korean LLM Leaderboard | 43.07 | Average Score (%) | 85.4 |
| Open LLM Leaderboard - GPQA | 8.5 | Score | 70.7 |
| Open LLM Leaderboard - IFEval | 45.17 | Score | 49.8 |
| Open LLM Leaderboard - BBH | 26.49 | Score | 41.5 |
| Open LLM Leaderboard - MMLU-Pro | 23.14 | Score | 39.3 |
| Open LLM Leaderboard - MuSR | 7.63 | Score | 35.6 |
| Open LLM Leaderboard - MATH Level 5 | 4.68 | Score | 26.5 |
Interactive version: theaggregate.ai/model?slug=llama-3-base-8b-sft-orpo · How It Works · Data refreshed daily, snapshot 2026-09-08.