High-performance inference framework for large language models, focusing on efficiency, flexibility, and availability.
Megatron-LM adapted for Puro-2B
High-performance inference framework for large language models, focusing on efficiency, flexibility, and availability.
Megatron-LM adapted for Puro-2B