Skip to content

[test]kimik3-fp4-b300-vllm-agentic-dspark: K=7, native offload #2419

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
xinli-sw wants to merge 3 commits into
base: main
Choose a base branch
from
+6 −0
Open

[test]kimik3-fp4-b300-vllm-agentic-dspark: K=7, native offload #2419

Changes from all commits
Commits
Show all changes
3 commits
Select commit Hold shift + click to select a range
4899e76
kimik3-fp4-b300-vllm-agentic-dspark: append perf-changelog entry / 追加…
xinli-sw Jul 31, 2026
91804e8
restore mtp recipes, switch 2xDEP2 prefill to 1xDEP4
xinli-sw Jul 31, 2026
179538b
Revert "restore mtp recipes, switch 2xDEP2 prefill to 1xDEP4"
xinli-sw Jul 31, 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
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 @@ -5355,3 +5355,9 @@
- "Apply the accuracy-gated Kimi-K2.5 MXFP4 settings: tuned AITER MXFP4 MoE, fused shared experts, FP8 KV cache, block size 16, 16384 batched tokens, 512 sequences, async scheduling, gpu-memory-utilization 0.85 (headroom for CUDA-graph capture on MI355X), and the AITER BF16 GEMM path"
- "Extend the TP4 and TP8 8k1k concurrency sweep from 64 to 128 (1k1k deprecated per #2263)"
pr-link: https://github.com/SemiAnalysisAI/InferenceX/pull/2213

- config-keys:
- kimik3-fp4-b300-vllm-agentic-dspark
description:
- "Update DSpark recipe: K=7 (AL 3.84), native offload, GPU-resident conc [1,2,4,8,12,14,16,18], offload conc [14,16,18,20,22,24], prefix-match-unit 64 (GPU-resident only), gsm8k eval."
pr-link: https://github.com/SemiAnalysisAI/InferenceX/pull/2419
Loading