Merge pull request #219 from imoneoi/docker

[docker] serving and ci workflow

Update README.md

Merge pull request #193 from imoneoi/moe-and-gemma-support

MoE and gemma support

Update README.md

Merge pull request #56 from imoneoi/3.5_mistral

3.5 mistral

Merge pull request #53 from imoneoi/metadata_fix

Metadata fix

[package] add flash-attn dependency

Merge branch '3.3_super'

[serving] fix stop token

set flashattention as optional dependency