comfyui-mcp/packs/ltx-2.3-xy-plot/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: ltx-2.3-xy-plot
display_name: LTX-2.3 — LTX-2.3 LoRA XY-plot grid
family: ltx
kind: txt2vid
description: >-
  LTX-2.3 LoRA XY-plot grid. LTX-2.3 (Lightricks, 22B GGUF) video generation, sliced from the
  LTX-2.3 ULTRA workflow into a standalone single-pipeline pack (no group toggles).
vram: "16GB+"
workflow: workflow.json
skill: ltxv2-video
launch_args: []
sources:
  model: https://huggingface.co/Lightricks/LTX-2
  mirror: https://huggingface.co/Aitrepreneur/FLX
notes:
  - Run the generated installer from your ComfyUI root.
  - Sliced + un-bypassed from the LTX-2.3 ULTRA workflow.
  - Needs the kornia fix (fix-ltxvideo-kornia.sh) if ComfyUI-LTXVideo's pyramid import breaks.
  - >-
    Bring your own trained LTX-2.3 LoRAs. This pack installs NO subject LoRAs —
    the two "easy loraNames" selectors ship the placeholders
    YOUR-LTX-LORA-a/b.safetensors, which are the two axes of the comparison grid.
    Drop your own .safetensors into models/loras/ and pick them there. The
    detailer and distill LoRAs the graph also loads ARE downloaded by this pack.
post_install:
  - Restart ComfyUI, load workflow.json.
  - >-
    Set the two "easy loraNames" nodes to your own trained LoRAs — until you do,
    they point at YOUR-LTX-LORA-a/b.safetensors placeholders and ComfyUI will
    report them missing.
# Declared external so packs:check-models knows they're intentionally not shipped.
# Placeholders, not real weights — see the "bring your own" note above (#1767).
external_models:
  - YOUR-LTX-LORA-a.safetensors
  - YOUR-LTX-LORA-b.safetensors