Skip to content

[AMD] retrigger dsv4-fp4-mi355x-atom benchmark sweep #1817

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

Merged
Oseltamivir merged 34 commits into from
Jun 18, 2026
+8 −0
Merged

[AMD] retrigger dsv4-fp4-mi355x-atom benchmark sweep #1817

Changes from all commits
Commits
Show all changes
34 commits
Select commit Hold shift + click to select a range
31b4fbe
[AMD] dsv4-fp4-mi355x-atom: enable DPA TBO at high concurrency, updat…
seungrokj Jun 12, 2026
c566e28
[AMD] perf-changelog: dsv4-fp4-mi355x-atom DPA TBO + image atom0.1.4
seungrokj Jun 12, 2026
7e1aa06
[AMD] perf-changelog: add PR link #1717
seungrokj Jun 12, 2026
65e0fa3
[AMD] dsv4_fp4_mi355x_atom.sh: disable prefix caching
seungrokj Jun 12, 2026
3f3560b
[AMD] dsv4-fp4-mi355x-atom: add max-model-len, eval context, extend c…
seungrokj Jun 12, 2026
c3b3289
[AMD] dsv4-fp4-mi355x-atom: narrow eval to single conc=1024 point, di…
seungrokj Jun 13, 2026
7ffa976
[AMD] dsv4_fp4_mi355x_atom.sh: add cudagraph-capture-sizes and max-nu…
seungrokj Jun 13, 2026
f2677b2
[AMD] dsv4-fp4-mi355x-atom: bump to nightly image, expand search spac…
seungrokj Jun 15, 2026
f5f0d66
[AMD] set GPU_MAX_HW_QUEUES=5 in dsv4_fp4_mi355x_atom.sh
seungrokj Jun 15, 2026
dc5b239
[AMD] dsv4-fp4-mi355x-atom: disable TBO, add TP4 rows for isl=8192, c…
seungrokj Jun 15, 2026
1dbf259
Merge branch 'main' into amd/dsv4_atom_0612
seungrokj Jun 15, 2026
9e18052
[AMD] dsv4_fp4_mi355x_atom.sh: quote SERVER_LOG variable
seungrokj Jun 16, 2026
c1812ed
[AMD] dsv4_fp4_mi355x_atom.sh: comment out dense cudagraph sizes
seungrokj Jun 16, 2026
28bdc6a
[AMD] dsv4_fp4_mi355x_atom.sh: fix --hf-overrides JSON escaping
seungrokj Jun 16, 2026
b36218e
[AMD] dsv4_fp4_mi355x_atom.sh: comment out dense cudagraph sizes
seungrokj Jun 16, 2026
fa47caf
[AMD] dsv4-fp4-mi355x-atom: expand search space, restore isl=1024 rows
seungrokj Jun 16, 2026
1022e0b
Merge branch 'main' into amd/dsv4_atom_0612
seungrokj Jun 16, 2026
af82c27
[AMD] perf-changelog: update dsv4-fp4-mi355x-atom image and search-sp…
seungrokj Jun 16, 2026
1300012
[AMD] dsv4_fp4_mi355x_atom.sh: restore sparse cudagraph capture sizes
seungrokj Jun 16, 2026
f56f877
[AMD] perf-changelog: revert dsv4-fp4-mi355x-atom image/search-space,…
seungrokj Jun 16, 2026
f7c9de8
Merge branch 'main' into amd/dsv4_atom_0612
seungrokj Jun 16, 2026
a4828cb
[AMD] perf-changelog: add dsv4-fp4-mi355x-sglang entry for PR #1762
seungrokj Jun 16, 2026
19b8757
update dsv4-fp4-mi355x-atom: bump image, enable TBO conditionally, fi…
seungrokj Jun 17, 2026
03aaa6b
expand dsv4-fp4-mi355x-atom search space: restore ISL1024 scenarios, …
seungrokj Jun 17, 2026
cf3962f
Merge branch 'main' into amd/dsv4_atom_0612
seungrokj Jun 17, 2026
421313c
Update perf-changelog.yaml
seungrokj Jun 17, 2026
ae77233
Update perf-changelog.yaml
seungrokj Jun 17, 2026
a8f6bd0
Update perf-changelog.yaml
seungrokj Jun 17, 2026
5fbd068
Update perf-changelog.yaml
seungrokj Jun 17, 2026
d080faa
update perf-changelog: move dsv4-fp4-mi355x-atom entry to end
seungrokj Jun 17, 2026
91f6277
narrow dsv4-fp4-mi355x-atom to DPA conc=256-2048 ISL8192, fix TBO bra…
seungrokj Jun 17, 2026
4364ef9
restore full dsv4-fp4-mi355x-atom search space: ISL1024 + ISL8192 TP4…
seungrokj Jun 17, 2026
52f9779
chore: retrigger dsv4 atom benchmark sweep
Oseltamivir Jun 18, 2026
dbc4d69
chore: retain PR 1717 sweep ancestry
Oseltamivir Jun 18, 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
8 changes: 8 additions & 0 deletions perf-changelog.yaml
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -3935,3 +3935,11 @@
- "Update ISL=8192 search-space: TP8-only from conc=4-64, DPA from conc=128-1024 (previously conc=1-64 and DPA conc=64-512)"
- "Update Applied TBO on high concurrencies"
pr-link: https://github.com/SemiAnalysisAI/InferenceX/pull/1717

- config-keys:
- dsv4-fp4-mi355x-atom
description:
- "Update image to rocm/atom:rocm7.2.4_ubuntu24.04_py3.12_pytorch_release_2.10.0_atom0.1.4_20260612"
- "Update ISL=8192 search-space: TP8-only from conc=4-64, DPA from conc=128-1024 (previously conc=1-64 and DPA conc=64-512)"
- "Update Applied TBO on high concurrencies"
pr-link: https://github.com/SemiAnalysisAI/InferenceX/pull/1717