Get up and running with Llama 3.3, DeepSeek-R1, Phi-4, Gemma 3, and other large language models.
[NeurIPS'23 Oral] Visual Instruction Tuning (LLaVA) built towards GPT-4V level capabilities and beyond.
