Skip to content

[AMD] Update MI355x Deepseek-R1 FP4 SGLang Image to v0.5.6.post1 #330

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
chunfangamd merged 7 commits into from
Dec 17, 2025
Merged

[AMD] Update MI355x Deepseek-R1 FP4 SGLang Image to v0.5.6.post1 #330

Show file tree
Hide file tree
Changes from all commits
Commits
Show all changes
7 commits
Select commit Hold shift + click to select a range
0d44123
Update amd-master.yaml
ppalanga Dec 15, 2025
0c9dcf5
Merge branch 'main' into ppalanga/update_image_to_0.5.6.post1
cquil11 Dec 15, 2025
d570a11
Merge branch 'main' into ppalanga/update_image_to_0.5.6.post1
cquil11 Dec 15, 2025
44aec22
Update perf-changelog.yaml
ppalanga Dec 15, 2025
1f98965
Update dsr1_fp4_mi355x_docker.sh
ppalanga Dec 15, 2025
2a5eebd
Merge branch 'main' into ppalanga/update_image_to_0.5.6.post1
cquil11 Dec 15, 2025
bf5ea3d
Update dsr1_fp4_mi355x_docker.sh
ppalanga Dec 15, 2025
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
2 changes: 1 addition & 1 deletion .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: rocm/7.0:rocm7.0_ubuntu_22.04_sgl-dev-v0.5.2-rocm7.0-mi35x-20250915
image: lmsysorg/sglang:v0.5.6.post1-rocm700-mi35x
model: amd/DeepSeek-R1-0528-MXFP4-Preview
model-prefix: dsr1
runner: mi355x
Expand Down
5 changes: 4 additions & 1 deletion benchmarks/dsr1_fp4_mi355x_docker.sh
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -11,6 +11,7 @@
# RESULT_FILENAME
# NUM_PROMPTS
export SGLANG_USE_AITER=1
export ROCM_QUICK_REDUCE_QUANTIZATION=INT4

PREFILL_SIZE=196608
if [[ "$ISL" == "8192" && "$OSL" == "1024" ]]; then
Expand All @@ -30,7 +31,9 @@ python3 -m sglang.launch_server --model-path=$MODEL --trust-remote-code \
--disable-radix-cache \
--num-continuous-decode-steps=4 \
--max-prefill-tokens=$PREFILL_SIZE \
--cuda-graph-max-bs=128 > $SERVER_LOG 2>&1 &
--cuda-graph-max-bs=128 \
--attention-backend aiter \
--kv-cache-dtype fp8_e4m3 > $SERVER_LOG 2>&1 &

SERVER_PID=$!

Expand Down
5 changes: 5 additions & 0 deletions perf-changelog.yaml
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -81,3 +81,8 @@
- Update vLLM image for NVIDIA configs from vLLM 0.11.0 to vLLM 0.11.2
- Adds kv-cache-dtype: fp8 to benchmarks/gptoss_fp4_b200_docker.sh
PR: https://github.com/InferenceMAX/InferenceMAX/pull/273
- config-keys:
- dsr1-fp4-mi355x-sglang
description: |
- Updating MI355x Deepseek-R1 FP4 SGLang Image to upstream v0.5.6.post1
PR: https://github.com/InferenceMAX/InferenceMAX/pull/330