Skip to content

[AMD] feat: Add mi355x distributed inference test CI workflow #409

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
cquil11 merged 7 commits into from
Jan 8, 2026
+601 −33
Merged

[AMD] feat: Add mi355x distributed inference test CI workflow #409

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
94e26f4
Revert "Revert "[AMD] Add mi355x distributed inference test CI workfl…
cquil11 Jan 7, 2026
b8f8004
add random range ratio that is appropriate (#402)
cquil11 Jan 7, 2026
afe471f
comment out 1k8k 8k1k
billishyahao Jan 8, 2026
5cd7366
change recipe back to upstream
billishyahao Jan 8, 2026
47df874
revert comment out 1k8k 8k1k
billishyahao Jan 8, 2026
1460803
Add empty line at the beginning of the script
cquil11 Jan 8, 2026
b0b9414
Merge branch 'main' into feat/mi355x-disagg-2
cquil11 Jan 8, 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
381 changes: 381 additions & 0 deletions .github/configs/amd-master.yaml
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -176,3 +176,384 @@ gptoss-fp4-mi355x-vllm:
- { tp: 1, conc-start: 4, conc-end: 128 }
- { tp: 4, conc-start: 4, conc-end: 4 }
- { tp: 8, conc-start: 4, conc-end: 8 }


dsr1-fp8-mi355x-sglang-disagg:
image: rocm/sgl-dev:sglang-0.5.6.post1-rocm700-mi35x-mori-1224
model: deepseek-ai/DeepSeek-R1-0528
model-prefix: dsr1
# NOTE: Until this cluster has more nodes, switch to a single runner because
# no oversubscription is needed.
runner: mi355x-amd_0
precision: fp8
framework: sglang-disagg
multinode: true
disagg: true
seq-len-configs:
- isl: 1024
osl: 1024
search-space:
# MTP configurations
# "Top of curve" (1 prefill worker at DEP8 and 1 decode worker at DEP16)
- spec-decoding: "mtp"
conc-list: [ 2048 ]
prefill:
num-worker: 1
tp: 1
ep: 8
dp-attn: true
additional-settings:
- "PREFILL_NODES=1"
decode:
num-worker: 1
tp: 1
ep: 16
dp-attn: true
additional-settings:
- "DECODE_NODES=2"
- "DECODE_MTP_SIZE=2"

# "Middle of curve" (1 prefill worker at DEP8 and 2 decode workers each at DEP8)
- spec-decoding: "mtp"
conc-list: [ 256, 512, 1024 ]
prefill:
num-worker: 1
tp: 1
ep: 8
dp-attn: true
additional-settings:
- "PREFILL_NODES=1"
decode:
num-worker: 2
tp: 1
ep: 8
dp-attn: true
additional-settings:
- "DECODE_NODES=2"
- "DECODE_MTP_SIZE=2"


# "Bottom of curve" (1 prefill worker at TEP8 and 2 decode workers at TEP8)
- spec-decoding: "mtp"
conc-list: [ 32, 64, 128 ]
prefill:
num-worker: 1
tp: 8
ep: 8
dp-attn: false
additional-settings:
- "PREFILL_NODES=1"

decode:
num-worker: 2
tp: 8
ep: 8
dp-attn: false
additional-settings:
- "DECODE_NODES=2"
- "DECODE_MTP_SIZE=2"

# non-MTP configurations
# "Top of curve" (1 prefill workers each at DEP8 and 1 decode workers at DEP16)
- spec-decoding: "none"
conc-list: [ 2048 ]
prefill:
num-worker: 1
tp: 1
ep: 8
dp-attn: true
additional-settings:
- "PREFILL_NODES=1"
decode:
num-worker: 1
tp: 1
ep: 16
dp-attn: true
additional-settings:
- "DECODE_NODES=2"
- "DECODE_MTP_SIZE=0"

# "Middle of curve" (1 prefill workers each at DEP8 and 2 decode workers at DEP8)
- spec-decoding: "none"
conc-list: [ 256, 512, 1024 ]
prefill:
num-worker: 1
tp: 1
ep: 8
dp-attn: true
additional-settings:
- "PREFILL_NODES=1"
decode:
num-worker: 2
tp: 1
ep: 8
dp-attn: true
additional-settings:
- "DECODE_NODES=2"
- "DECODE_MTP_SIZE=0"


# "Bottom of curve" (1 prefill worker at TEP8 and 2 decode workers at TEP8)
- spec-decoding: "none"
conc-list: [ 32, 64, 128 ]
prefill:
num-worker: 1
tp: 8
ep: 8
dp-attn: false
additional-settings:
- "PREFILL_NODES=1"

decode:
num-worker: 2
tp: 8
ep: 8
dp-attn: false
additional-settings:
- "DECODE_NODES=2"
- "DECODE_MTP_SIZE=0"

- isl: 8192
osl: 1024
search-space:
# MTP configurations
# "Top of curve" (1 prefill worker at DEP8 and 1 decode worker at DEP16)
- spec-decoding: "mtp"
conc-list: [ 2048 ]
prefill:
num-worker: 1
tp: 1
ep: 8
dp-attn: true
additional-settings:
- "PREFILL_NODES=1"
decode:
num-worker: 1
tp: 1
ep: 16
dp-attn: true
additional-settings:
- "DECODE_NODES=2"
- "DECODE_MTP_SIZE=2"

# "Middle of curve" (1 prefill worker at DEP8 and 2 decode workers each at DEP8)
- spec-decoding: "mtp"
conc-list: [ 256, 512, 1024 ]
prefill:
num-worker: 1
tp: 1
ep: 8
dp-attn: true
additional-settings:
- "PREFILL_NODES=1"
decode:
num-worker: 2
tp: 1
ep: 8
dp-attn: true
additional-settings:
- "DECODE_NODES=2"
- "DECODE_MTP_SIZE=2"


# "Bottom of curve" (1 prefill worker at TEP8 and 2 decode workers at TEP8)
- spec-decoding: "mtp"
conc-list: [ 32, 64, 128 ]
prefill:
num-worker: 1
tp: 8
ep: 8
dp-attn: false
additional-settings:
- "PREFILL_NODES=1"

decode:
num-worker: 2
tp: 8
ep: 8
dp-attn: false
additional-settings:
- "DECODE_NODES=2"
- "DECODE_MTP_SIZE=2"

# non-MTP configurations
# "Top of curve" (1 prefill workers each at DEP8 and 1 decode workers at DEP16)
- spec-decoding: "none"
conc-list: [ 2048 ]
prefill:
num-worker: 1
tp: 1
ep: 8
dp-attn: true
additional-settings:
- "PREFILL_NODES=1"
decode:
num-worker: 1
tp: 1
ep: 16
dp-attn: true
additional-settings:
- "DECODE_NODES=2"
- "DECODE_MTP_SIZE=0"

# "Middle of curve" (1 prefill workers each at DEP8 and 2 decode workers at DEP8)
- spec-decoding: "none"
conc-list: [ 256, 512, 1024 ]
prefill:
num-worker: 1
tp: 1
ep: 8
dp-attn: true
additional-settings:
- "PREFILL_NODES=1"
decode:
num-worker: 2
tp: 1
ep: 8
dp-attn: true
additional-settings:
- "DECODE_NODES=2"
- "DECODE_MTP_SIZE=0"

# "Bottom of curve" (1 prefill worker at TEP8 and 2 decode workers at TEP8)
- spec-decoding: "none"
conc-list: [ 32, 64, 128 ]
prefill:
num-worker: 1
tp: 8
ep: 8
dp-attn: false
additional-settings:
- "PREFILL_NODES=1"

decode:
num-worker: 2
tp: 8
ep: 8
dp-attn: false
additional-settings:
- "DECODE_NODES=2"
- "DECODE_MTP_SIZE=0"


- isl: 1024
osl: 8192
search-space:
# MTP configurations
# "Top of curve" (1 prefill workers each at DEP8 and 2 decode workers at DEP8)
- spec-decoding: "mtp"
conc-list: [ 2048 ]
prefill:
num-worker: 1
tp: 1
ep: 8
dp-attn: true
additional-settings:
- "PREFILL_NODES=1"
decode:
num-worker: 1
tp: 1
ep: 16
dp-attn: true
additional-settings:
- "DECODE_NODES=2"
- "DECODE_MTP_SIZE=2"


# "Middle of curve" (1 prefill worker at DEP8 and 2 decode workers each at DEP8)
- spec-decoding: "mtp"
conc-list: [ 256, 512, 1024 ]
prefill:
num-worker: 1
tp: 1
ep: 8
dp-attn: true
additional-settings:
- "PREFILL_NODES=1"
decode:
num-worker: 2
tp: 1
ep: 8
dp-attn: true
additional-settings:
- "DECODE_NODES=2"
- "DECODE_MTP_SIZE=2"


# "Bottom of curve" (1 prefill worker at TEP8 and 2 decode workers at TEP8)
- spec-decoding: "mtp"
conc-list: [ 32, 64, 128 ]
prefill:
num-worker: 1
tp: 8
ep: 8
dp-attn: false
additional-settings:
- "PREFILL_NODES=1"

decode:
num-worker: 2
tp: 8
ep: 8
dp-attn: false
additional-settings:
- "DECODE_NODES=2"
- "DECODE_MTP_SIZE=2"

# non-MTP configurations
# "Top of curve" (1 prefill workers each at DEP8 and 1 decode workers at DEP16)
- spec-decoding: "none"
conc-list: [ 2048 ]
prefill:
num-worker: 1
tp: 1
ep: 8
dp-attn: true
additional-settings:
- "PREFILL_NODES=1"
decode:
num-worker: 1
tp: 1
ep: 16
dp-attn: true
additional-settings:
- "DECODE_NODES=2"
- "DECODE_MTP_SIZE=0"

# "Middle of curve" (1 prefill workers each at DEP8 and 2 decode workers at DEP8)
- spec-decoding: "none"
conc-list: [ 256, 512, 1024 ]
prefill:
num-worker: 1
tp: 1
ep: 8
dp-attn: true
additional-settings:
- "PREFILL_NODES=1"
decode:
num-worker: 2
tp: 1
ep: 8
dp-attn: true
additional-settings:
- "DECODE_NODES=2"
- "DECODE_MTP_SIZE=0"

# "Bottom of curve" (1 prefill worker at TEP8 and 2 decode workers at TEP8)
- spec-decoding: "none"
conc-list: [ 32, 64, 128 ]
prefill:
num-worker: 1
tp: 8
ep: 8
dp-attn: false
additional-settings:
- "PREFILL_NODES=1"
decode:
num-worker: 2
tp: 8
ep: 8
dp-attn: false
additional-settings:
- "DECODE_NODES=2"
- "DECODE_MTP_SIZE=0"
Comment on lines +192 to +559

@gemini-code-assist gemini-code-assist Bot Jan 8, 2026

Copy link
Copy Markdown
Contributor

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

medium

The search-space configuration is duplicated three times for different isl/osl combinations. This makes the file very long and hard to maintain. You can use YAML anchors and aliases to define the search-space once and reuse it. This will significantly reduce the file size and improve readability.

For example:

# At the top level of your YAML, define an anchor
x-dsr1-search-space: &dsr1_search_space
  # MTP configurations
  # "Top of curve" (1 prefill worker at DEP8 and 1 decode worker at DEP16)
  - spec-decoding: "mtp"
    conc-list: [ 2048 ]
    # ... rest of the search space items

ds1-fp8-mi355x-sglang-disagg:
  # ... other properties
  seq-len-configs:
  - isl: 1024
    osl: 1024
    search-space: *dsr1_search_space # Reuse with an alias
  - isl: 8192
    osl: 1024
    search-space: *dsr1_search_space # Reuse with an alias
  - isl: 1024
    osl: 8192
    search-space: *dsr1_search_space # Reuse with an alias

Loading