kimik3-fp4-mi355x-vllm-agentic-mtp: add LMCache DRAM KV-offload arm - #2583

Closed

sammshen wants to merge 13 commits into

SemiAnalysisAI:mainfrom

sammshen:kimik3-mi355x-lmcache

Commits on Aug 12, 2026

kimik3-fp4-mi355x-vllm-agentic-mtp: add LMCache DRAM KV-offload arm

sammshen
authored and
functionstackx
committed
055dfc2
perf-changelog: select kimik3-fp4-mi355x-vllm-agentic-mtp LMCache arm

sammshen
authored and
functionstackx
committed
328836b
Move the LMCache arm to a dedicated config key at conc 4/8/16
sammshen
committed
4584d37
Lower the lmcache key's dram-utilization to fit the shm-backed L1
sammshen
committed
aba14d1
Match the LMCache chunk size to this stack's 1536-token block size
sammshen
committed
6b95b7b
Raise the LMCache chunk size to 3072 for the KDA state group
sammshen
committed
a8cda1a
Pin the LMCache MP server to the lmcache_driven transfer path
sammshen
committed
afee680
Hold LMCache L1 read locks for the job duration
sammshen
committed
20b4fda
Revert "Hold LMCache L1 read locks for the job duration"
sammshen
committed
3c1908b

Commits on Aug 13, 2026

Merge branch 'main' into kimik3-mi355x-lmcache
ApostaC
committed
d671e53
trigger lmcache 0.5.4rc2
ApostaC
committed
0811fb0
update lmcache configus
ApostaC
committed
ac6ad7d
update conc
ApostaC
committed
e890a2b