4/6 Fine-Tuning Index shows #OSS #LLMs outperforming GPT-4 on 85% of tested tasks

→ View original post on X — @predibase