adapter-serving

Here are 2 public repositories matching this topic...

prosoftdev999 / lora-slot-lineage

ML inference forensics benchmark for reconstructing LoRA adapter lineage across reusable slots, reloads, evictions, reassignments, and runtime restarts.

python docker distributed-systems machine-learning checkpoint state-reconstruction lora observability lineage model-serving peft event-replay llm ml-inference adapter-serving

Updated Sep 19, 2026
Python

JohnScheuer / multi-lora-serving-sim

Star

Discrete-event simulation of multi-LoRA adapter serving strategies on a single GPU — comparing naive swap, hot-set preloading, and batch-by-adapter under variable VRAM pressure and arrival rates.

python simulation scheduling inference pytorch memory-management lora lru-cache llm-serving multi-lora vram-management adapter-serving s-lora mlsystems

Updated Jul 18, 2026
Python

Add this topic to your repo

To associate your repository with the adapter-serving topic, visit your repo's landing page and select "manage topics."

Learn more