LaunchDevelopers
28 days ago
hf-mem adds MoE memory breakdown

Hugging Face
@huggingfaceThe AI community building the future. https://t.co/TpiXQMQ9rZ
NYC and Paris and πhuggingface.co

Hugging Face
@huggingface
RT @alvarobartt: Latest `hf-mem` now breaks down Mixture-of-Experts (MoE) memory estimations into base weights, routed experts, and KV cachβ¦
Β·
28 days ago