Skip to content

[AMD][MI35X] Serve Qwen3.5 MXFP4-AttnFP8-V2 on the MI355X SGLang single-node arms #2695

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
yichiche wants to merge 3 commits into
base: main
Choose a base branch
from
Draft

[AMD][MI35X] Serve Qwen3.5 MXFP4-AttnFP8-V2 on the MI355X SGLang single-node arms #2695

Show file tree
Hide file tree
Changes from all commits
Commits
Show all changes
3 commits
Select commit Hold shift + click to select a range
3bce80b
[AMD][MI35X] Serve Qwen3.5 MXFP4-AttnFP8-V2 on the MI355X SGLang sing…
yichiche Aug 23, 2026
72cbc41
[AMD][MI35X] Export SGLANG_USE_AITER_FP8_PER_TOKEN for the AttnFP8-V2…
yichiche Aug 23, 2026
2fdc164
Merge branch 'main' into amd/qwen3.5-fp4-attnfp8-mi355x-sglang
yichiche Aug 25, 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
1 change: 1 addition & 0 deletions benchmarks/single_node/fixed_seq_len/qwen3.5_fp4_mi355x.sh
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -21,6 +21,7 @@ export SGLANG_USE_AITER=1
export SGLANG_USE_AITER_UNIFIED_ATTN=1
export AITER_FLYDSL_FORCE=1
export SGLANG_MAMBA_SSM_DTYPE=bfloat16
export SGLANG_USE_AITER_FP8_PER_TOKEN=1

SERVER_LOG=/workspace/server.log
MEM_FRAC_STATIC=${MEM_FRAC_STATIC:-0.8}
Expand Down
1 change: 1 addition & 0 deletions benchmarks/single_node/fixed_seq_len/qwen3.5_fp4_mi355x_mtp.sh
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -21,6 +21,7 @@ export SGLANG_USE_AITER=1
export SGLANG_USE_AITER_UNIFIED_ATTN=1
export AITER_FLYDSL_FORCE=1
export SGLANG_MAMBA_SSM_DTYPE=bfloat16
export SGLANG_USE_AITER_FP8_PER_TOKEN=1

SERVER_LOG=/workspace/server.log
MEM_FRAC_STATIC=${MEM_FRAC_STATIC:-0.8}
Expand Down
4 changes: 2 additions & 2 deletions configs/amd-master.yaml
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -314,7 +314,7 @@ qwen3.5-fp8-mi355x-sglang-disagg:

qwen3.5-fp4-mi355x-sglang:
image: lmsysorg/sglang-rocm:v0.5.16-rocm720-mi35x-20260730
model: amd/Qwen3.5-397B-A17B-MXFP4
model: amd/Qwen3.5-397B-A17B-MXFP4-AttnFP8-V2
Comment thread
cursor[bot] marked this conversation as resolved.
model-prefix: qwen3.5
runner: mi355x
precision: fp4
Expand Down Expand Up @@ -346,7 +346,7 @@ qwen3.5-fp4-mi355x-atom:

qwen3.5-fp4-mi355x-sglang-mtp:
image: lmsysorg/sglang-rocm:v0.5.16-rocm720-mi35x-20260730
model: amd/Qwen3.5-397B-A17B-MXFP4
model: amd/Qwen3.5-397B-A17B-MXFP4-AttnFP8-V2
model-prefix: qwen3.5
runner: mi355x
precision: fp4
Expand Down
10 changes: 10 additions & 0 deletions perf-changelog.yaml
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -6409,3 +6409,13 @@
- "Require dcgm-power telemetry on the GLM-5.2 H200 disaggregated AgentX recipe."
- "Enable dcgm-power telemetry on the DSV4 H200 aggregated AgentX recipe."
pr-link: https://github.com/SemiAnalysisAI/InferenceX/pull/2684

- config-keys:
- qwen3.5-fp4-mi355x-sglang
- qwen3.5-fp4-mi355x-sglang-mtp
scenario-type:
- fixed-seq-len
description:
- "Switch the served checkpoint from amd/Qwen3.5-397B-A17B-MXFP4 to amd/Qwen3.5-397B-A17B-MXFP4-AttnFP8-V2 on both arms."
- "Export SGLANG_USE_AITER_FP8_PER_TOKEN=1 on both arms, as the checkpoint's published SGLang recipe requires."
pr-link: https://github.com/SemiAnalysisAI/InferenceX/pull/2695
Loading