Multi-GPU

Shard-Any-LLMs

An automated model-sharding workflow for distributing large language models across multiple GPUs.