About Inferwatch
Inferwatch is an MCP server in the Blockchain category: real-time and historical metrics for locally served LLMs (Ollama and vLLM) — dashboard, HTTP API and MCP server. One process, one SQLite file, never in the request path.
Install
git clone https://github.com/floatsmyboat/inferwatchThis server has no ConduID identity, so agent calls to it are not receipted. Pin the version you install and review the source before granting it credentials.
Tools (9)
cache_statusOllama prompt-cache occupancy and eviction pressure, plus live KV usage
Parameters not published.
client_statswho is calling, for which models, at what context size
Parameters not published.
concurrencyrequests in flight at once, and how close to slot capacity
Parameters not published.
get_eventscold loads, evictions, truncations, warnings
Parameters not published.
gpu_statusper-device util/VRAM/temp/power
Parameters not published.
healthis collection working, is debug logging on
Parameters not published.
image_failureswhat failed, by node class, and what never reached a backend
Parameters not published.
image_generationsrecent generations with every model the workflow loaded
Parameters not published.
image_summarySwarmUI/ComfyUI throughput, durations, models, backends
Parameters not published.
Ask AI
Ask AI about Inferwatch
Powered by Claude · Grounded in docs
Security checks
6 checks have not been run yet.