Local Image MCP
Related Servers
Alternatives to Local Image MCP
No user-submitted related servers found.
Related Servers
- AlicenseNot gradedqualityCmaintenanceEnables driving a self-hosted ComfyUI instance over stdio MCP for image and song generation, workflow execution, job and queue management, and local model, asset, and workflow administration. Every submit is priced against the machine's own VRAM meter first, so work the GPU cannot pay for is refused before it is sent, and an unreadable meter also refuses rather than permitting.Business Source 1.1
- AlicenseNot gradedqualityDmaintenanceLocal-first MCP server for controlling a loopback ComfyUI instance from same-machine or authenticated LAN clients.MIT
- FlicenseAqualityDmaintenanceMinimal MCP server that wraps ComfyUI HTTP routes enabling LLM clients to inspect server status, models, features, node metadata, submit workflows, manage queue/history, interrupt runs, and upload images.11-
- FlicenseNot gradedqualityCmaintenanceWindows-native MCP server for a local ComfyUI portable package. It enables listing, validating, and running ComfyUI workflows via the loopback HTTP API.-
- AlicenseNot gradedqualityDmaintenanceEnables AI image generation from text prompts by wrapping local ComfyUI as a Streamable HTTP MCP server, supporting synchronous generation, queue management, and history tracking.1MIT
- AlicenseNot gradedqualityCmaintenanceEnables MCP clients to connect to and interact with ComfyUI for workflow execution and image generation.Apache 2.0
TDQS
Scored across 4 tools
Each tool has a distinct role: generate submits a text-to-image job, get_result polls an existing job, status is read-only health inspection, and free_vram manages GPU memory. The only mild overlap is between generate and get_result, but the generate description explicitly tells the agent to switch to get_result on timeout, which resolves the boundary.
Three of four names follow a verb_noun / verb pattern (generate, free_vram, get_result), while 'status' is a bare noun. The deviation is minor and the intent of every name is immediately readable.
Four tools is compact and every tool earns its place: job submission, job polling, environment diagnostics, and VRAM release. Nothing feels padded or missing at the count level for a single-model local image generator.
The surface covers the core lifecycle (submit → poll → retrieve image) plus health checks and VRAM management, which is solid for text-to-image only. Gaps remain for image-to-image/upscale, seed or negative-prompt control, and cancelling a queued job, though agents can mostly work around these.