sgl-project/sglang
[Feature] Support encoder models for flash_infer backend
クローズ
#6,050 opened on 2025/05/06
good first issue
Repository metrics
- Stars
- (28,442 個のスター)
- PR merge metrics
- (平均マージ 2d 1h) (30d で 1,000 merged PRs)
説明
Checklist
- 1. If the issue you raised is not a feature but a question, please raise a discussion at https://github.com/sgl-project/sglang/discussions/new/choose Otherwise, it will be closed.
- 2. Please use English, otherwise it will be closed.
Motivation
@DavidBao03 hi, currently only support encoder model with torch_native attn backend and triton attn backend. Other attn backend is not supported yet.
Originally posted by @woodx9 in https://github.com/sgl-project/sglang/issues/4887#issuecomment-2847365703
Related resources
No response