hf-mem 0.1.1

A CLI to estimate inference memory requirements for Hugging Face models, written in Python

hf-mem is an experimental CLI to estimate inference memory requirements for Hugging Face models, written in Python. hf-mem is lightweight, only depends on httpx. It's recommended to run with uv for a better experience.

hf-mem lets you estimate the inference requirements to run any model from the Hugging Face Hub, including Transformers, Diffusers and Sentence Transformers models, as well as any model that contains Safetensors compatible weights.

Read more information about hf-mem in this short-form post.

Usage

uvx hf-mem --model-id MiniMaxAI/MiniMax-M2

uvx hf-mem --model-id Qwen/Qwen-Image

References

• Safetensors Metadata parsing
• usgraphics - TR-100 Machine Report