MiuLab/Taiwan-LLM
Taiwanese LLM project providing traditional Mandarin language models including Llama-3-Taiwan with fine-tuning and evaluation infrastructure.
Velocity · 7d
+1.4
★ / day
Trend
→steady
star history
Taiwan-LLM (TAME) is a collection of large language models specialized in traditional Mandarin and Taiwanese culture. The project provides model weights on Hugging Face, fine-tuning configurations using Axolotl, and evaluation benchmarks including chatbot arena rankings. It partners with healthcare, technology, and academic institutions to develop domain-specific Taiwanese language models.