Skip to content

Bump vLLM version for DSV4 B300 disagg #1952

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
adibarra merged 2 commits into from
Jun 29, 2026
Merged

Bump vLLM version for DSV4 B300 disagg #1952

Show file tree
Hide file tree
Changes from 1 commit
Commits
Show all changes
2 commits
Select commit Hold shift + click to select a range
a89c646
Bump vLLM version for DSV4 B300 disagg
xinli-sw Jun 29, 2026
127b530
perf-changelog: fix PR link for B300 vLLM bump to #1952
xinli-sw Jun 29, 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
2 changes: 1 addition & 1 deletion .github/configs/nvidia-master.yaml
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -9622,7 +9622,7 @@ dsv4-fp4-gb200-dynamo-sglang-mtp:
ep: 16
dp-attn: true
dsv4-fp4-b300-dynamo-vllm:
image: vllm/vllm-openai:v0.20.1
image: vllm/vllm-openai:v0.23.0
model: deepseek-ai/DeepSeek-V4-Pro
model-prefix: dsv4
runner: b300
Expand Down
6 changes: 3 additions & 3 deletions ...arks/multi_node/srt-slurm-recipes/vllm/deepseek-v4/8k1k/disagg-b300-high-tpt-megamoe.yaml
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -4,7 +4,7 @@ name: "svf-vllm-disagg-b300-high-tpt-megamoe"
# one full 8-GPU B300 node, plus a dedicated NATS/etcd infra node.
model:
path: "deepseek-v4-pro"
container: "vllm/vllm-openai:v0.20.1"
container: "vllm/vllm-openai:v0.23.0"
precision: "fp4"

dynamo:
Expand Down Expand Up @@ -124,7 +124,7 @@ identity:
repo: "deepseek-ai/DeepSeek-V4-Pro"
revision: "0366e4e064385807ea86b088a5c6c878ff23343b"
container:
image: "vllm/vllm-openai:v0.20.1"
image: "vllm/vllm-openai:v0.23.0"
frameworks:
dynamo: "1.2.0.dev20260426"
vllm: "0.20.1"
vllm: "0.23.0"
6 changes: 3 additions & 3 deletions benchmarks/multi_node/srt-slurm-recipes/vllm/deepseek-v4/8k1k/disagg-b300-low-latency.yaml
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -4,7 +4,7 @@ name: "svf-vllm-disagg-b300-low-latency"
# one full 8-GPU B300 node, plus a dedicated NATS/etcd infra node.
model:
path: "deepseek-v4-pro"
container: "vllm/vllm-openai:v0.20.1"
container: "vllm/vllm-openai:v0.23.0"
precision: "fp4"

dynamo:
Expand Down Expand Up @@ -118,7 +118,7 @@ benchmark:

identity:
container:
image: "vllm/vllm-openai:v0.20.1"
image: "vllm/vllm-openai:v0.23.0"
frameworks:
dynamo: "1.2.0.dev20260426"
vllm: "0.20.1"
vllm: "0.23.0"
6 changes: 3 additions & 3 deletions ...arks/multi_node/srt-slurm-recipes/vllm/deepseek-v4/8k1k/disagg-b300-low-middle-curve.yaml
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -4,7 +4,7 @@ name: "svf-vllm-disagg-b300-low-middle-curve"
# one full 8-GPU B300 node, plus a dedicated NATS/etcd infra node.
model:
path: "deepseek-v4-pro"
container: "vllm/vllm-openai:v0.20.1"
container: "vllm/vllm-openai:v0.23.0"
precision: "fp4"

dynamo:
Expand Down Expand Up @@ -119,7 +119,7 @@ benchmark:

identity:
container:
image: "vllm/vllm-openai:v0.20.1"
image: "vllm/vllm-openai:v0.23.0"
frameworks:
dynamo: "1.2.0.dev20260426"
vllm: "0.20.1"
vllm: "0.23.0"
6 changes: 3 additions & 3 deletions ...marks/multi_node/srt-slurm-recipes/vllm/deepseek-v4/8k1k/disagg-b300-max-tpt-megamoe.yaml
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -4,7 +4,7 @@ name: "svf-vllm-disagg-b300-max-tpt-megamoe"
# one full 8-GPU B300 node, plus a dedicated NATS/etcd infra node.
model:
path: "deepseek-v4-pro"
container: "vllm/vllm-openai:v0.20.1"
container: "vllm/vllm-openai:v0.23.0"
precision: "fp4"

dynamo:
Expand Down Expand Up @@ -124,7 +124,7 @@ identity:
repo: "deepseek-ai/DeepSeek-V4-Pro"
revision: "0366e4e064385807ea86b088a5c6c878ff23343b"
container:
image: "vllm/vllm-openai:v0.20.1"
image: "vllm/vllm-openai:v0.23.0"
frameworks:
dynamo: "1.2.0.dev20260426"
vllm: "0.20.1"
vllm: "0.23.0"
6 changes: 3 additions & 3 deletions ...rks/multi_node/srt-slurm-recipes/vllm/deepseek-v4/8k1k/disagg-b300-mid-curve-megamoe.yaml
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -4,7 +4,7 @@ name: "svf-vllm-disagg-b300-mid-curve-megamoe"
# one full 8-GPU B300 node, plus a dedicated NATS/etcd infra node.
model:
path: "deepseek-v4-pro"
container: "vllm/vllm-openai:v0.20.1"
container: "vllm/vllm-openai:v0.23.0"
precision: "fp4"

dynamo:
Expand Down Expand Up @@ -124,7 +124,7 @@ identity:
repo: "deepseek-ai/DeepSeek-V4-Pro"
revision: "0366e4e064385807ea86b088a5c6c878ff23343b"
container:
image: "vllm/vllm-openai:v0.20.1"
image: "vllm/vllm-openai:v0.23.0"
frameworks:
dynamo: "1.2.0.dev20260426"
vllm: "0.20.1"
vllm: "0.23.0"
6 changes: 6 additions & 0 deletions perf-changelog.yaml
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -4302,3 +4302,9 @@
- "Update the MiniMax-M3 MXFP8 MI355X vLLM image from vllm/vllm-openai-rocm:minimax-m3 to vllm/vllm-openai-rocm:nightly-3f5a1e1733200760169ff31ebe60a271072b199e."
- "Benchmark serving flags and TP/EP/DP-attention search space are unchanged."
pr-link: https://github.com/SemiAnalysisAI/InferenceX/pull/1942

- config-keys:
- dsv4-fp4-b300-dynamo-vllm
description:
- "Update the DeepSeek-V4-Pro B300 disaggregated Dynamo-vLLM benchmark to the vllm/vllm-openai:v0.23.0 image"
pr-link: https://github.com/SemiAnalysisAI/InferenceX/pull/1899

Check warning on line 4310 in perf-changelog.yaml

View check run for this annotation

Claude / Claude Code Review

perf-changelog pr-link points to wrong PR 

The new `perf-changelog.yaml` entry for `dsv4-fp4-b300-dynamo-vllm` (lines 4306-4310) sets `pr-link` to PR #1899, but that is the B200 PR being mirrored — the link should point to this PR (#1952). Every adjacent entry in the file links to the PR that introduced it (#1921/#1939/#1941/#1942 on lines 4282/4290/4297/4304), so this looks like a copy-paste artifact from the description ("Mirrors the B200 image bump from #1899"). Fix by changing the `pr-link` to `https://github.com/SemiAnalysisAI/Infer

@claude claude Bot Jun 29, 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.

🟡 The new perf-changelog.yaml entry for dsv4-fp4-b300-dynamo-vllm (lines 4306-4310) sets pr-link to PR #1899, but that is the B200 PR being mirrored — the link should point to this PR (#1952). Every adjacent entry in the file links to the PR that introduced it (#1921/#1939/#1941/#1942 on lines 4282/4290/4297/4304), so this looks like a copy-paste artifact from the description ("Mirrors the B200 image bump from #1899"). Fix by changing the pr-link to https://github.com/SemiAnalysisAI/InferenceX/pull/1952.

Extended reasoning...

What the bug is

The new entry appended to perf-changelog.yaml at lines 4306-4310 records the B300 vLLM image bump from v0.20.1 to v0.23.0 for dsv4-fp4-b300-dynamo-vllm. Its pr-link is set to https://github.com/SemiAnalysisAI/InferenceX/pull/1899, but this PR is #1952 — "Bump vLLM version for DSV4 B300 disagg." PR #1899 is the prior B200 image bump that this PR is mirroring, as the PR description itself states ("Mirrors the B200 image bump from #1899 for the B300 equivalents").

How it manifests / what the impact is

perf-changelog.yaml is a curated record of configuration changes, where each entry's pr-link is meant to point reviewers to the PR that introduced that specific change. The convention is unambiguous across the surrounding entries:

Each of these links to the PR that actually added the entry. The new B300 entry breaks the convention: a future reader tracing the B300 image bump via the changelog would be redirected to PR #1899 (the B200 bump for a different image and recipes), not the PR that performed the B300 bump. This defeats the entire point of the pr-link field for this entry.

Why existing code doesn't prevent it

perf-changelog.yaml is plain YAML metadata — there is no schema validation or automated check that the pr-link value matches the PR adding the entry. The mistake therefore slips through anything short of human review, which is exactly what this comment is for.

Step-by-step proof

  1. PR metadata shows this is PR Bump vLLM version for DSV4 B300 disagg #1952, title "Bump vLLM version for DSV4 B300 disagg."
  2. The PR description explicitly says: "Mirrors the B200 image bump from Bump vLLM version for DSV4 B200 disagg #1899 for the B300 equivalents."
  3. The diff against perf-changelog.yaml appends one new entry whose pr-link field is https://github.com/SemiAnalysisAI/InferenceX/pull/1899.
  4. The four immediately preceding entries (still visible in the same file, lines ~4279-4304) each link to their own introducing PR ([NV]Add Qwen3.5-397B-A17B-NVFP4 GB300 disagg multinode SGLang via Dynamo #1921, [codex] add MiniMax M3 FP4 MI355X vLLM MTP benchmark #1939, [codex] update MiniMax M3 FP8 MI355X vLLM MTP image #1941, [codex] update MiniMax M3 FP8 MI355X vLLM image #1942) — establishing the convention.
  5. Conclusion: the value #1899 is a copy-paste from the description's source PR, not the PR introducing this entry. It should be #1952.

How to fix

Change the last line of the new entry from

  pr-link: https://github.com/SemiAnalysisAI/InferenceX/pull/1899

to

  pr-link: https://github.com/SemiAnalysisAI/InferenceX/pull/1952

No runtime behavior is affected — this is purely a changelog-traceability correction, but worth catching before merge while the fix is trivial.

Loading