Bo-Eval v1 Collection Bo-Eval v1 is the premier open-source benchmark for Tibetan AI. It rigorously evaluates LLM reasoning and STT accuracy across diverse dialects. • 2 items • Updated about 9 hours ago