Skip to content

[AgentX] MiniMax-M3.1-Flash GB300 SGLang: store DSpark draft KV in FP8 / [AgentX] MiniMax-M3.1-Flash GB300 SGLang：DSpark draft KV 改用 FP8 #4600

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

Open
zijiexia wants to merge 2 commits into
base: main
Choose a base branch
from
+12 −0
Open

[AgentX] MiniMax-M3.1-Flash GB300 SGLang: store DSpark draft KV in FP8 / [AgentX] MiniMax-M3.1-Flash GB300 SGLang：DSpark draft KV 改用 FP8 #4600

Show file tree
Hide file tree
Changes from all commits
Commits
Show all changes
2 commits
Select commit Hold shift + click to select a range
38dc886
Store DSpark draft KV in FP8 on the GB300 SGLang AgentX curve
zijiexia Oct 28, 2026
3b96f33
Set the changelog PR link
zijiexia Oct 28, 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: 2 additions & 0 deletions ...marks/single_node/srt-slurm-recipes/minimaxm31flash/sglang/gb300-fp4-mtp/agentic-tp2.yaml
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -33,6 +33,8 @@ base:
speculative-algorithm: DSPARK
speculative-dspark-block-size: 7
speculative-use-block-verification: true
# FP8 draft KV, as vLLM's DSpark drafter uses; bf16 costs ~10 GB of HBM per rank.
speculative-draft-kv-cache-dtype: fp8_e4m3
cuda-graph-backend-prefill: breakable
tokenizer-worker-num: 8
watchdog-timeout: 3600
Expand Down
2 changes: 2 additions & 0 deletions ...enchmarks/single_node/srt-slurm-recipes/minimaxm31flash/sglang/gb300-fp4-mtp/agentic.yaml
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -33,6 +33,8 @@ base:
speculative-algorithm: DSPARK
speculative-dspark-block-size: 7
speculative-use-block-verification: true
# FP8 draft KV, as vLLM's DSpark drafter uses; bf16 costs ~10 GB of HBM per rank.
speculative-draft-kv-cache-dtype: fp8_e4m3
Comment on lines +36 to +37

@claude claude Bot Oct 28, 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.

🔴 After merging, the GB300 MiniMax-M3.1-Flash SGLang AgentX curve runs its DSpark draft below shipped precision, which CONTRIBUTING.md#draft-model-precision forbids. speculative-draft-kv-cache-dtype: fp8_e4m3 targets only the draft. No target kv-cache-dtype is set, so target KV stays at the bf16 default. Fix: drop the draft-only override in both recipes and their changelog line, or use one upstream-supported KV dtype for target and draft alike. Same pattern at 2 sites (gb300-fp4-mtp/agentic.yaml:37, gb300-fp4-mtp/agentic-tp2.yaml:37).

Why this was flagged

The base agg role args of agentic.yaml:37 and agentic-tp2.yaml:37 now set speculative-draft-kv-cache-dtype: fp8_e4m3. Neither recipe sets a target kv-cache-dtype. The PR's own changelog entry (perf-changelog.yaml, new block) says this replaces "the bf16 default", so the effective draft KV precision drops below the pinned image's default for this checkpoint. On the base branch the draft KV runs at the shipped bf16 default and is compliant. CONTRIBUTING.md lists "Dtype or KV-cache dtype overrides aimed at the draft that lower its precision below the framework's default for that checkpoint" as forbidden. It says matching an upstream recipe (here, vLLM's DSpark drafter) and passing evals or synthetic AL do not exempt this. The allowed case is FP8 draft KV inherited from an FP8 target KV, which does not apply here. Every point except agentic-low-latency.yaml in config key minimaxm31flash-fp4-gb300-sglang-agentic-dspark would publish non-compliant throughput gained by cheapening the draft.

Verification: agentic.yaml:36-37 and agentic-tp2.yaml:36-37 add speculative-draft-kv-cache-dtype: fp8_e4m3 to the base agg args, and no SGLang minimaxm31flash recipe sets kv-cache-dtype. So the effective draft KV precision drops from the pinned image's default (bf16) to FP8, which CONTRIBUTING.md lines 86-87 forbid. On the base branch the draft runs at the default bf16 KV and is compliant.

cuda-graph-backend-prefill: breakable
tokenizer-worker-num: 8
watchdog-timeout: 3600
Expand Down
8 changes: 8 additions & 0 deletions inferencex-e2e/perf-changelog.yaml
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -11489,3 +11489,11 @@
- "Update the MI355X DeepSeek-V4-Pro-0813 SGLang MoRI disaggregated AgentX serving image from lmsysorg/sglang-rocm:v0.5.22-rocm720-mi35x-20261019 to lmsysorg/sglang-rocm:v0.5.22-rocm10-mi35x-20261026, the latest published MI35x ROCm 10 nightly (digest sha256:edabad83728ab04e3eae39f315eb16261eeb64dbdb051112bd22fc88560b60c4)."
- "Preserve all eight concurrency points (1, 16, 32, 48, 192, 256, 384, 512), prefill/decode topology, DSpark settings, UMBP offload, MoRI transfer, and benchmark client image. The workload binder supplies the serving image to every selected recipe variant."
pr-link: https://github.com/SemiAnalysisAI/InferenceX/pull/4571

- config-keys:
- minimaxm31flash-fp4-gb300-sglang-agentic-dspark
description:
- "Store the DSpark draft KV cache in FP8 (speculative-draft-kv-cache-dtype fp8_e4m3) on every point except the batch-size-1 MonoKernel Low-Latency point, instead of the bf16 default, matching the vLLM DSpark drafter; this frees HBM and HiCache host pool space for target KV."
- "Rerun the whole curve so it is measured on one recipe revision."
pr-link: https://github.com/SemiAnalysisAI/InferenceX/pull/4600
scenario-type: [agentic-coding]
Loading