Skip to content

[Klaud Cold] Update dsv4-fp4-b300-dynamo-sglang SGLang image to v0.5.19-cu130 (digest-pinned) and refresh the Dynamo pin / 将 dsv4-fp4-b300-dynamo-sglang 的 SGLang 镜像更新为 v0.5.19-cu130（按 digest 固定）并刷新 Dynamo 提交 #2896

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

Closed
Klaud-Cold wants to merge 1 commit into from
Closed

[Klaud Cold] Update dsv4-fp4-b300-dynamo-sglang SGLang image to v0.5.19-cu130 (digest-pinned) and refresh the Dynamo pin / 将 dsv4-fp4-b300-dynamo-sglang 的 SGLang 镜像更新为 v0.5.19-cu130（按 digest 固定）并刷新 Dynamo 提交 #2896

Show file tree
Hide file tree
Changes from all commits
Commits
Show all changes
1 commit
Select commit
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
10 changes: 4 additions & 6 deletions ...arks/multi_node/srt-slurm-recipes/sglang/deepseek-v4/8k1k/disagg-b300-1p1d-dep4-dep8.yaml
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -2,11 +2,11 @@ name: "disagg-b300-1p1d-dep4-dep8"

model:
path: "deepseek-v4-pro"
container: "lmsysorg/sglang:nightly-dev-cu13-20260721-8905cbd4"
container: "lmsysorg/sglang:v0.5.19-cu130@sha256:d6e7288627be8b02be88e4bba38e73f6d50e2826869f753c13a4c4385ab3eda9"
precision: "fp4"

dynamo:
hash: "41882ae9b07232eed4850fb1daf8c958abb2556a"
hash: "b1c5147f6c9eb036d98e0e25e2a980375f491e78"
install: true

sbatch_directives:
Expand Down Expand Up @@ -46,8 +46,6 @@ backend:
SGLANG_DSV4_REASONING_EFFORT: "max"
SGLANG_OPT_SWA_SPLIT_LEAF_ON_INSERT: "1"
SGLANG_OPT_DEEPGEMM_MEGA_MOE_NUM_MAX_TOKENS_PER_RANK: "8192"
SGLANG_OPT_DEEPGEMM_MEGA_MOE_USE_FP4_ACTS: "1"
SGLANG_OPT_DEEPGEMM_MEGA_MOE_USE_MXF4_KIND: "1"
SGLANG_OPT_USE_ONLINE_COMPRESS: "1"
NCCL_MNNVL_ENABLE: "1"
NCCL_CUMEM_ENABLE: "1"
Expand All @@ -69,8 +67,6 @@ backend:
SGLANG_DSV4_REASONING_EFFORT: "max"
SGLANG_OPT_SWA_SPLIT_LEAF_ON_INSERT: "1"
SGLANG_OPT_DEEPGEMM_MEGA_MOE_NUM_MAX_TOKENS_PER_RANK: "1280"
SGLANG_OPT_DEEPGEMM_MEGA_MOE_USE_FP4_ACTS: "1"
SGLANG_OPT_DEEPGEMM_MEGA_MOE_USE_MXF4_KIND: "1"
SGLANG_OPT_USE_ONLINE_COMPRESS: "1"
NCCL_MNNVL_ENABLE: "1"
NCCL_CUMEM_ENABLE: "1"
Expand Down Expand Up @@ -99,6 +95,7 @@ backend:

enable-dp-attention: true
moe-a2a-backend: "megamoe"
enable-w4a4-mxfp4-megamoe: true
moe-dense-tp-size: 1

disaggregation-mode: "prefill"
Expand All @@ -118,6 +115,7 @@ backend:
stream-interval: 60

moe-a2a-backend: "megamoe"
enable-w4a4-mxfp4-megamoe: true
moe-dense-tp-size: 1

disaggregation-mode: "decode"
Expand Down
4 changes: 2 additions & 2 deletions ...hmarks/multi_node/srt-slurm-recipes/sglang/deepseek-v4/8k1k/disagg-b300-1p1d-tp4-tp4.yaml
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -2,11 +2,11 @@ name: "disagg-b300-1p1d-tp4-tp4"

model:
path: "deepseek-v4-pro"
container: "lmsysorg/sglang:nightly-dev-cu13-20260721-8905cbd4"
container: "lmsysorg/sglang:v0.5.19-cu130@sha256:d6e7288627be8b02be88e4bba38e73f6d50e2826869f753c13a4c4385ab3eda9"
precision: "fp4"

dynamo:
hash: "41882ae9b07232eed4850fb1daf8c958abb2556a"
hash: "b1c5147f6c9eb036d98e0e25e2a980375f491e78"
install: true

sbatch_directives:
Expand Down
10 changes: 4 additions & 6 deletions ...arks/multi_node/srt-slurm-recipes/sglang/deepseek-v4/8k1k/disagg-b300-2p1d-dep4-dep8.yaml
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -2,11 +2,11 @@ name: "disagg-b300-2p1d-dep4-dep8"

model:
path: "deepseek-v4-pro"
container: "lmsysorg/sglang:nightly-dev-cu13-20260721-8905cbd4"
container: "lmsysorg/sglang:v0.5.19-cu130@sha256:d6e7288627be8b02be88e4bba38e73f6d50e2826869f753c13a4c4385ab3eda9"
precision: "fp4"

dynamo:
hash: "41882ae9b07232eed4850fb1daf8c958abb2556a"
hash: "b1c5147f6c9eb036d98e0e25e2a980375f491e78"
install: true

sbatch_directives:
Expand Down Expand Up @@ -46,8 +46,6 @@ backend:
SGLANG_DSV4_REASONING_EFFORT: "max"
SGLANG_OPT_SWA_SPLIT_LEAF_ON_INSERT: "1"
SGLANG_OPT_DEEPGEMM_MEGA_MOE_NUM_MAX_TOKENS_PER_RANK: "8192"
SGLANG_OPT_DEEPGEMM_MEGA_MOE_USE_FP4_ACTS: "1"
SGLANG_OPT_DEEPGEMM_MEGA_MOE_USE_MXF4_KIND: "1"
SGLANG_OPT_USE_ONLINE_COMPRESS: "1"
NCCL_MNNVL_ENABLE: "1"
NCCL_CUMEM_ENABLE: "1"
Expand All @@ -69,8 +67,6 @@ backend:
SGLANG_DSV4_REASONING_EFFORT: "max"
SGLANG_OPT_SWA_SPLIT_LEAF_ON_INSERT: "1"
SGLANG_OPT_DEEPGEMM_MEGA_MOE_NUM_MAX_TOKENS_PER_RANK: "1280"
SGLANG_OPT_DEEPGEMM_MEGA_MOE_USE_FP4_ACTS: "1"
SGLANG_OPT_DEEPGEMM_MEGA_MOE_USE_MXF4_KIND: "1"
SGLANG_OPT_USE_ONLINE_COMPRESS: "1"
NCCL_MNNVL_ENABLE: "1"
NCCL_CUMEM_ENABLE: "1"
Expand Down Expand Up @@ -99,6 +95,7 @@ backend:

enable-dp-attention: true
moe-a2a-backend: "megamoe"
enable-w4a4-mxfp4-megamoe: true
moe-dense-tp-size: 1

disaggregation-mode: "prefill"
Expand All @@ -118,6 +115,7 @@ backend:
stream-interval: 60

moe-a2a-backend: "megamoe"
enable-w4a4-mxfp4-megamoe: true
moe-dense-tp-size: 1

disaggregation-mode: "decode"
Expand Down
10 changes: 4 additions & 6 deletions ...arks/multi_node/srt-slurm-recipes/sglang/deepseek-v4/8k1k/disagg-b300-4p1d-dep4-dep8.yaml
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -2,11 +2,11 @@ name: "disagg-b300-4p1d-dep4-dep8"

model:
path: "deepseek-v4-pro"
container: "lmsysorg/sglang:nightly-dev-cu13-20260721-8905cbd4"
container: "lmsysorg/sglang:v0.5.19-cu130@sha256:d6e7288627be8b02be88e4bba38e73f6d50e2826869f753c13a4c4385ab3eda9"
precision: "fp4"

dynamo:
hash: "41882ae9b07232eed4850fb1daf8c958abb2556a"
hash: "b1c5147f6c9eb036d98e0e25e2a980375f491e78"
install: true

sbatch_directives:
Expand Down Expand Up @@ -46,8 +46,6 @@ backend:
SGLANG_DSV4_REASONING_EFFORT: "max"
SGLANG_OPT_SWA_SPLIT_LEAF_ON_INSERT: "1"
SGLANG_OPT_DEEPGEMM_MEGA_MOE_NUM_MAX_TOKENS_PER_RANK: "8192"
SGLANG_OPT_DEEPGEMM_MEGA_MOE_USE_FP4_ACTS: "1"
SGLANG_OPT_DEEPGEMM_MEGA_MOE_USE_MXF4_KIND: "1"
SGLANG_OPT_USE_ONLINE_COMPRESS: "1"
NCCL_MNNVL_ENABLE: "1"
NCCL_CUMEM_ENABLE: "1"
Expand All @@ -69,8 +67,6 @@ backend:
SGLANG_DSV4_REASONING_EFFORT: "max"
SGLANG_OPT_SWA_SPLIT_LEAF_ON_INSERT: "1"
SGLANG_OPT_DEEPGEMM_MEGA_MOE_NUM_MAX_TOKENS_PER_RANK: "1280"
SGLANG_OPT_DEEPGEMM_MEGA_MOE_USE_FP4_ACTS: "1"
SGLANG_OPT_DEEPGEMM_MEGA_MOE_USE_MXF4_KIND: "1"
SGLANG_OPT_USE_ONLINE_COMPRESS: "1"
NCCL_MNNVL_ENABLE: "1"
NCCL_CUMEM_ENABLE: "1"
Expand Down Expand Up @@ -99,6 +95,7 @@ backend:

enable-dp-attention: true
moe-a2a-backend: "megamoe"
enable-w4a4-mxfp4-megamoe: true
moe-dense-tp-size: 1

disaggregation-mode: "prefill"
Expand All @@ -118,6 +115,7 @@ backend:
stream-interval: 60

moe-a2a-backend: "megamoe"
enable-w4a4-mxfp4-megamoe: true
moe-dense-tp-size: 1

disaggregation-mode: "decode"
Expand Down
10 changes: 4 additions & 6 deletions ...arks/multi_node/srt-slurm-recipes/sglang/deepseek-v4/8k1k/disagg-b300-6p1d-dep4-dep8.yaml
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -2,11 +2,11 @@ name: "disagg-b300-6p1d-dep4-dep8"

model:
path: "deepseek-v4-pro"
container: "lmsysorg/sglang:nightly-dev-cu13-20260721-8905cbd4"
container: "lmsysorg/sglang:v0.5.19-cu130@sha256:d6e7288627be8b02be88e4bba38e73f6d50e2826869f753c13a4c4385ab3eda9"
precision: "fp4"

dynamo:
hash: "41882ae9b07232eed4850fb1daf8c958abb2556a"
hash: "b1c5147f6c9eb036d98e0e25e2a980375f491e78"
install: true

sbatch_directives:
Expand Down Expand Up @@ -46,8 +46,6 @@ backend:
SGLANG_DSV4_REASONING_EFFORT: "max"
SGLANG_OPT_SWA_SPLIT_LEAF_ON_INSERT: "1"
SGLANG_OPT_DEEPGEMM_MEGA_MOE_NUM_MAX_TOKENS_PER_RANK: "8192"
SGLANG_OPT_DEEPGEMM_MEGA_MOE_USE_FP4_ACTS: "1"
SGLANG_OPT_DEEPGEMM_MEGA_MOE_USE_MXF4_KIND: "1"
SGLANG_OPT_USE_ONLINE_COMPRESS: "1"
NCCL_MNNVL_ENABLE: "1"
NCCL_CUMEM_ENABLE: "1"
Expand All @@ -69,8 +67,6 @@ backend:
SGLANG_DSV4_REASONING_EFFORT: "max"
SGLANG_OPT_SWA_SPLIT_LEAF_ON_INSERT: "1"
SGLANG_OPT_DEEPGEMM_MEGA_MOE_NUM_MAX_TOKENS_PER_RANK: "1280"
SGLANG_OPT_DEEPGEMM_MEGA_MOE_USE_FP4_ACTS: "1"
SGLANG_OPT_DEEPGEMM_MEGA_MOE_USE_MXF4_KIND: "1"
SGLANG_OPT_USE_ONLINE_COMPRESS: "1"
NCCL_MNNVL_ENABLE: "1"
NCCL_CUMEM_ENABLE: "1"
Expand Down Expand Up @@ -99,6 +95,7 @@ backend:

enable-dp-attention: true
moe-a2a-backend: "megamoe"
enable-w4a4-mxfp4-megamoe: true
moe-dense-tp-size: 1

disaggregation-mode: "prefill"
Expand All @@ -118,6 +115,7 @@ backend:
stream-interval: 60

moe-a2a-backend: "megamoe"
enable-w4a4-mxfp4-megamoe: true
moe-dense-tp-size: 1

disaggregation-mode: "decode"
Expand Down
2 changes: 1 addition & 1 deletion configs/nvidia-master.yaml
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -5233,7 +5233,7 @@ dsv4-fp4-b300-dynamo-vllm:
# runners/launch_b300-nv.sh. Prefill workers are TP4/DEP4 (two workers per
# 8-GPU B300 node); the decode worker is a single DEP8 node.
dsv4-fp4-b300-dynamo-sglang:
image: lmsysorg/sglang:nightly-dev-cu13-20260721-8905cbd4
image: lmsysorg/sglang:v0.5.19-cu130@sha256:d6e7288627be8b02be88e4bba38e73f6d50e2826869f753c13a4c4385ab3eda9
model: deepseek-ai/DeepSeek-V4-Pro
model-prefix: dsv4
runner: b300
Expand Down