Commit 2fccf4a
SGLang H20 Inference Blog (#209)
* feat: Initial commit for sglang-ant-group
* upd
* SBO
* DeepX
* fix: title
* Challenge and Solution.Deployment
* fix: deployment
* fix: deployment
* fix: deployment
* Performance: Computation
* EPLB
* fix: title
* add intro
* comment
* refactor: challenge
* comment
* refine: EPLB and computation
* fix: TBO
* fix: DeepX
* prefill
* refine: challenges with H20
* refine: Observability
* refine: Deployment Strategy
* add image
* refine: Performance
* refine: SBO
* upd
* foramt
* format
* format
* upd
* upd
* refine: Performance
* add figures
* fix: figure name
* refine: Environment in Decode
* refine: Environment in Decode
* refine: Observation + Solution in Optimizations
* refine: Observation + Solution in Optimizations
* add new figure
* upd
* format
* upd
* add Prefill in Performance
* add Prefill in Performance
* upd sbo.
* Acknowledgements
* author
* Open Source
* fix: Performance-Prefill
* fix: Expert Affinity EPLB link
* upd
* refine: simplify benefit in Solution
* refine: remove Related PRs
* figure: update
* figure: update
* upd
* figure: update
* upd
* figure: update
* figure: add preview image
* figure: deploy
* figure: deploy
* fix: title
* fix: logo
* fix: logo
* fix: image size
* fix: format
* fix: figure
* fix: format
* fix: deploy.svg
* fix: prefill_perf.png
* fix: logo.svg
* fix: antgroup/sglang repo
* fix: Acknowledgements
* figure: update
* update
* update
* add Conclusion and update
* fix
* polish
* update figure
* upd
* change release date
* fix CR
---------
Co-authored-by: 墨纭 <moyun.zty@antgroup.com>
Co-authored-by: 昶知 <eric.hc@antgroup.com>
Co-authored-by: 剑川 <jianchuan.gys@antgroup.com>1 parent 1241987 commit 2fccf4a
14 files changed
Lines changed: 331 additions & 0 deletions
Large diffs are not rendered by default.
Loading
Loading
Loading
Loading
Loading
Loading
Loading
Loading
Loading
0 commit comments