/usr/bin/lmcache

[2026-11-08 04:34:15,296] LMCache INFO: torch_dev=StubCPUDevice(device_type=cpu), torch_device_type=cpu (_device_detect.py:311:lmcache.v1.platform._device_detect)

 _     __  __    ____           _
| |   |  \/  |  / ___|__ _  ___| |__   ___      LMCache v0.5.7
| |   | |\/| | | |   / _` |/ __| '_ \ / _ \     Website:  https://lmcache.ai/
| |___| |  | | | |__| (_| | (__| | | |  __/     Recipes:  https://docs.lmcache.ai/recipes
|_____|_|  |_|  \____\__,_|\___|_| |_|\___|     LinkedIn: https://www.linkedin.com/company/lmcache-lab
Set LMCACHE_DISABLE_BANNER=1 to hide this banner.

usage: lmcache [-h]
               {bench,coordinator,describe,kvcache,memory,mock,ping,query,quota,server,tool,trace}
               ...

LMCache — KV cache management for LLM serving

positional arguments:
  {bench,coordinator,describe,kvcache,memory,mock,ping,query,quota,server,tool,trace}
    bench               Run sustained performance benchmarks.
    coordinator         Launch the LMCache mp coordinator (HTTP).
    describe            Show detailed status of a running LMCache service.
    kvcache             Manage KV cache state.
    memory              Launch the memory orchestrator: the allocation and readiness authority for
                        one shared memory region.
    mock                Run a mock command (example/test).
    ping                Ping LMCache or vLLM server (liveness check).
    query               Run one inference request and report metrics.
    quota               Manage per-salt cache quotas.
    server              Launch the LMCache request server and HTTP API.
    tool                Run offline analysis tools.
    trace               Inspect and replay LMCache trace files.

options:
  -h, --help            show this help message and exit