comfyui-mcp/packs/z-image-base-combo/pack.yaml

Local-first, agent-native control plane for ComfyUI — MCP server + autonomous sidebar agent that drives your live graph in natural language on ANY LLM: Claude/ChatGPT/Gemini on your subscription (no API key), free local models via Ollama (fully offline),

name: z-image-base-combo
display_name: Z-Image Base — Ultra Combo (txt2img + img2img)
family: z-image
kind: combo
description: >-
  Z-Image Base (Alibaba Tongyi, 6B S3-DiT base model) Ultra Combo pipeline that
  chains the BASE and distilled TURBO models for a two-stage generate-then-refine
  pass, covering both text-to-image and image-to-image. Uses a Qwen text encoder
  (not CLIP/T5), GGUF-quantized to run in <8GB VRAM, with an 8-step distill LoRA,
  a Detailer Daemon detail pass, and optional film-grain / laplacian-sharpen
  post-processing. Single-purpose combo graph sliced from the z-image-base monolith
  (no group toggles).
vram: "<8GB"
workflow: workflow.json
skill: z-image-txt2img
launch_args: []
sources:
  model: https://huggingface.co/Aitrepreneur/FLX
  mirror: https://huggingface.co/Aitrepreneur/FLX
notes:
  - Run the generated installer from your ComfyUI root (folder with custom_nodes/
    and models/).
  - Sliced + un-bypassed from the monolith's "Z-IMAGE ULTRA COMBO TEXT TO IMAGE"
    and "Z-IMAGE ULTRA COMBO IMAGE TO IMAGE" pipelines.
  - The combo chains both the base (z_image) and turbo (z_image_turbo) GGUF models.
  - The z-image VAE is saved as z-image-ae.safetensors (its weights differ from
    Flux's ae.safetensors despite the same size).
post_install:
  - Restart ComfyUI (or ComfyUI-Manager -> Install missing custom nodes for any
    node Python deps, incl. the Detailer Daemon).
  - Load workflow.json.