deployedbyai
Follow

The Deploy Log | Models and APIs | Free, no signup

vLLM transformers backend

Hugging Face | In Edition 42,

The call on this one

DEPLOYif you serve Hugging Face models with vLLM

SOURCE Hugging Face, Native-speed vLLM transformers modeling backend

The transformers vLLM backend now meets or beats native throughput on Qwen3 4B dense, 32B dense, and 235B-parameter FP8 MoE models, using torch.fx static analysis and ast source rewriting to apply inference-specific layer fusions at runtime.

Added to The Deploy Log , updated . Read Edition 42, the edition that carried it.

70 rows from Hugging Face in The Deploy Log are dated on or before this one, this one included.

The same day from Hugging Face: LeRobot v0.6.0 (0 days apart). After it: Nunchaku Lite in Diffusers, 12 days later.

|

One line every time AI actually ships, as it ships, 24/7. Follow the wire: Threads | Bluesky

One line every time AI actually ships, around the clock, free on Threads and Bluesky. Every one is in The Deploy Log with its source.

The Saturday edition is the week in one, free to read on the site, and by email if you want it: get it by email.