Skip to content

Docker image NVIDIA GH200 machines - optimized for vllm serving and hf trainer finetuning

License

Notifications You must be signed in to change notification settings

abacusai/gh200-llm

Folders and files

NameName
Last commit message
Last commit date

Latest commit

 

History

43 Commits
 
 
 
 
 
 
 
 
 
 

Repository files navigation

GH200s

The new NVIDIA GH200 chips hold a lot of promise because of their high CPU->GPU memory bandwidth (7x the H100 DGX setup). Node to node communication along is supported by InfiniBand HDR 200Gbps non-blocking and ConnectX-7 NIC ratio for GPUDirect RDMA. NVIDIA has published some impressive raw numbers.

We ran some initial real world tests comparing a new batch of GH200 96GB GPU memory, 480GB node memory nodes that we got from Lambda Labs. The single node GH200 machines are an awesome deal and hosting ~70B models on a single node in fp8 mode is a great use case.

This repo maintains a ready to use docker image which works on H100s and GH200s (multi architecture) with the latest versions of VLLM, XFormers and Flash Attention to easily be able to serve large models and finetune 8B ones.

Docker image is published at: ghcr.io/abacusai/gh200-llm/llm-train-serve:latest

About

Docker image NVIDIA GH200 machines - optimized for vllm serving and hf trainer finetuning

Resources

License

Stars

Watchers

Forks

Releases

No releases published

Packages

 
 
 

Contributors 3

  •  
  •  
  •