GitLab

Enable 720p model inference on low-spec GPUs/CPUs and accelerate T5/CLIP quantized models with vLLM operators

Co-authored-by: Linboyan-trc <1584340372@qq.com>

* add step & cfg distillation wan model

* add step & cfg distillation wan model

* bug fixed

* reconstruct quantization and fix memory leak bug.

* Support lazy load inference.

* reconstruct quantization

* Fix hunyuan bugs

* deleted tmp file

---------
Co-authored-by: root <root@pt-c0b333b3a1834e81a0d4d5f412c6ffa1-worker-0.pt-c0b333b3a1834e81a0d4d5f412c6ffa1.ns-devsft-3460edd0.svc.cluster.local>
Co-authored-by: gushiqiao <gushqiaio@sensetime.com>
Co-authored-by: gushiqiao <gushiqiao@sensetime.com>

* split dit server from default runner

* split dit server from default runner

* update loading functions

* simplify loader functions and runner functions

* simplify code && split dit service

* simplify code && split dit service

* support split server for cogvideox

* clear code.

* fix offload extra latency in the first step by pre-allocating pinned memory

* pre-commit

---------
Co-authored-by: “de1star” <“843414674@qq.com”>

* support prompt enhancer server

* bugs fixed

* finished prompt enhancer service

* add load_transformer methods for split server

* add service utils

* [feature] support split servers

* async offload & context4debug

* offload ratio

* Merge branch 'main' into xinchi/fix_offload

* adding offload ratio

* pre-commit

---------
Co-authored-by: “de1star” <“843414674@qq.com”>