mcpcost

io.github.zc4578980-tech/local-gpu-imagegen

Local GPU Imagegen

Is the Local GPU Imagegen MCP server down? No. It answered initialize and tools/list in the probe of 2026-10-31.

MCP-first control plane for ComfyUI workflows with explicit approvals and no silent downloads.

alive probed 2026-11-15 · unchanged since 2026-10-31

17tools
8.5ktokens in Claude Opus 5.5
4.2 %of Claude's 200k window
3.4×the median server
976 msto initialize
9 msto list tools
Run it
uvx local-gpu-imagegen@0.9.0 (stdio, probed in a container with the network off)
Host
pypi.org
Reports itself as
local-gpu-imagegen 0.9.0
Capabilities
tools
Instructions
none
Of a 1M window
0.8 % of a 1M context window
Count
8.5k tokens in Claude Opus 5.5, exact, measured 2026-10-17. With the Claude Code prefix mcp__…__ on each tool name: 8.7k. The o200k_base estimate is 4.4k. Counted again when the tool definitions change. Method.
Repository
https://github.com/zc4578980-tech/local-gpu-imagegen
Website
https://github.com/zc4578980-tech/local-gpu-imagegen
Registry
active, updated 2026-08-08 · registry entry

Most expensive tools

The Claude column scales the o200k estimate of each tool by this server's ratio of exact to estimate, 1.95×.

Tool≈ Claude tokenso200k estimateShare of this server
local_gpu_record_review1 85895322 %
local_gpu_set_model_trust1 57580819 %
local_gpu_start_run88945610 %
local_gpu_generate_image6863528 %
local_gpu_recommend_models5402776 %

Compare this server with others.

More servers

Other registry entries on pypi.org, the ones next to this name; all of them.

Servers that mention imagegen.

Badge for the README

mcpcost: alive

The result of the latest probe, the tool count and the token cost. The feed has an entry for every probe that changed them.

Atom feed

A mail when this server changes

Stops answering, comes back, tool set or cost changes. One confirmation mail first, a stop link in every mail.

For the publisher