Skip to content

DSR1-FP4 MI355x SGLang: Add EP configurations #1811

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Draft
ppalanga wants to merge 3 commits into
base: main
Choose a base branch
from
+22 −6
Draft

DSR1-FP4 MI355x SGLang: Add EP configurations #1811

Changes from 1 commit
Commits
Show all changes
3 commits
Select commit Hold shift + click to select a range
016686f
Update amd-master.yaml
ppalanga Jun 17, 2026
98e80e6
Update dsr1_fp4_mi355x.sh
ppalanga Jun 17, 2026
008eeaa
Update perf-changelog.yaml
ppalanga Jun 17, 2026
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
15 changes: 10 additions & 5 deletions .github/configs/amd-master.yaml
View file
Open in desktop
Original file line number Diff line number Diff line change
@@ -1,5 +1,5 @@
dsr1-fp4-mi355x-sglang:
image: lmsysorg/sglang:v0.5.12-rocm700-mi35x
image: lmsysorg/sglang:v0.5.13-rocm700-mi35x
model: amd/DeepSeek-R1-0528-MXFP4-Preview
model-prefix: dsr1
runner: mi355x
Expand All @@ -11,13 +11,18 @@ dsr1-fp4-mi355x-sglang:
- isl: 1024
osl: 1024
search-space:
- { tp: 4, conc-start: 4, conc-end: 64 }
- { tp: 8, conc-start: 4, conc-end: 64 }
- { tp: 4, conc-start: 4, conc-end: 256 }
- { tp: 8, conc-start: 4, conc-end: 256 }
- { tp: 4, ep: 4, conc-start: 4, conc-end: 256 }
- { tp: 8, ep: 8, conc-start: 4, conc-end: 256 }
Comment thread
cursor[bot] marked this conversation as resolved.
- isl: 8192
osl: 1024
search-space:
- { tp: 4, conc-start: 4, conc-end: 64 }
- { tp: 8, conc-start: 4, conc-end: 64 }
- { tp: 4, conc-start: 4, conc-end: 256 }
- { tp: 8, conc-start: 4, conc-end: 256 }
- { tp: 4, ep: 4, conc-start: 4, conc-end: 256 }
- { tp: 8, ep: 8, conc-start: 4, conc-end: 256 }

# Agentic-coding sweep commented out for this image-bump PR — the
# 10-conc agentic matrix amplifies sweep cost and the bump validation
# only needs the fixed-seq-len throughput shape. Re-enable once the
Expand Down