inference-cache plugin
Inference KV-cache and transport suite — LMCache multiprocess (MP) standalone-server mode (DaemonSet + Deployment K8s pattern, ZMQ connector, L1 + L2 NIXL/POSIX/GDS/HF3FS/fs/s3/mooncake adapters) and NVIDIA NIXL transfer library (UCX/GDS/Mooncake/libfabric/HF3FS/S3 plugins, agent API, telemetry) used by Dynamo/vLLM/SGLang. Pairs with the vllm-caching skill in the `vllm` suite and sglang-hicache in the `sglang` suite.
How to install the inference-cache plugin
Run both commands inside Claude Code. The first adds the marketplace once; the second installs the plugin.
/plugin marketplace add air-gapped/skills
/plugin install inference-cache@air-gapped-marketplaceWhat's inside inference-cache
2 parts read from the plugin's own files.
Skills 2
Claude loads it when the task matches- lmcache-mp
Enables a standalone LMCache server in a separate process or pod that vLLM connects to via ZMQ for shared KV-cache and process isolation.
- nvidia-nixl
Provides a transport abstraction for point-to-point KV-cache and tensor transfers in distributed inference environments.
Using inference-cache in Claude Code
Skills need no command. Claude reads them when your request matches what they describe.
Update, turn off or remove
- Update the plugin (shell)
- claude plugin update inference-cache@air-gapped-marketplace
- Refresh the marketplace
- /plugin marketplace update air-gapped-marketplace
- Turn it off for now
- /plugin disable inference-cache@air-gapped-marketplace
- Remove it
- /plugin uninstall inference-cache@air-gapped-marketplace
Skills in inference-cache
- lmcache-mpair-gapped/skills
Enables a standalone LMCache server in a separate process or pod that vLLM connects to via ZMQ for shared KV-cache and process isolation.
8 files6stars - nvidia-nixlair-gapped/skills
Provides a transport abstraction for point-to-point KV-cache and tensor transfers in distributed inference environments.
12 files6stars
inference-cache plugin questions
How do I install the inference-cache plugin?
In Claude Code run /plugin marketplace add air-gapped/skills, then /plugin install inference-cache@air-gapped-marketplace.
What does the inference-cache plugin include?
inference-cache bundles 2 skills.
How do I update or remove the inference-cache plugin?
Update it with claude plugin update inference-cache@air-gapped-marketplace in your shell, or from the Installed tab of /plugin. Remove it with /plugin uninstall inference-cache@air-gapped-marketplace.
What license is the inference-cache plugin under?
MIT, as stated in the plugin's manifest.