ComfyUI From Zero - Part 1 (free) - downloads

The files at the top are the ones mentioned in the video. The Extras folder is a bonus.

ComfyUI-From-Zero-Part1-slides.pdf
    Every slide of Part 1 (123 pages), to flip back through.

comfyui-cheatsheet.pdf
    The two-page cheat sheet from Module 5: the files you download and the nodes that load them on page 1; the
    settings, the canvas, the hardware and where to download what on page 2. Print it or keep it on a second monitor.

workflow-01-first-image.json
    Module 4. Drag it onto the ComfyUI canvas and read the yellow note first.
    It needs one file: z_image_turbo_aio.safetensors in ComfyUI/models/checkpoints/ (then press R).
    The download link is in the note, and under "Model downloads" at the end of this file.
    Settings are for a turbo model: steps 8, CFG 1. The seed is fixed at 42, so you get the same picture every time you run it.

workflow-03-broken.json
    The spicy homework from the last module: broken in three places on purpose. Two stop it running, the third lets
    it run and gives you something that looks wrong. Read the error before you change anything.

workflow-03-answers.txt
workflow-03-fixed.json
    The answers to that homework: for each of the three breaks, what you see, why it happens and the fix, plus the
    repaired workflow. Try it yourself first.

help-request-template.txt
    Module 9. Copy it, fill in the five lines, paste it wherever you ask for help.

Extras - workflows from the on-screen demos/
    The small workflows used for the demonstrations in the video (the seven-node starter, the cable rules, bypass
    vs mute, save vs preview, and one workflow per error from Module 9, each broken on purpose). Most use the model
    from the video, waiIllustriousSDXL_v170.safetensors; swap in any checkpoint you have.
    course_p1_cat.png goes in ComfyUI/input/ for the Load Image demo. What each one loads:

P1 - Bypass vs mute.json
    model   waiIllustriousSDXL_v170.safetensors  ->  models/checkpoints
    model   your_style_lora.safetensors  ->  models/loras  (a stand-in name: pick your own file)

P1 - Cable rules.json
    model   waiIllustriousSDXL_v170.safetensors  ->  models/checkpoints

P1 - Checkpoint vs three loaders.json
    model   waiIllustriousSDXL_v170.safetensors  ->  models/checkpoints
    model   qwen_image_2.1_int8_convrot.safetensors  ->  models/diffusion_models
    model   qwen3vl_8b_int8_convrot.safetensors  ->  models/text_encoders
    model   qwen_image_2.1_vae_bf16.safetensors  ->  models/vae

P1 - Error - missing custom node.json
    model   waiIllustriousSDXL_v170.safetensors  ->  models/checkpoints
    nodes   a pack that is missing on purpose (that is the lesson)

P1 - Error - prompt has no outputs.json
    model   waiIllustriousSDXL_v170.safetensors  ->  models/checkpoints

P1 - Error - required input is missing.json
    model   waiIllustriousSDXL_v170.safetensors  ->  models/checkpoints

P1 - Error - shape mismatch.json
    model   qwen_image_2.1_vae_bf16.safetensors  ->  models/vae

P1 - Error - value not in list.json
    model   definitely_not_a_real_model.safetensors  ->  models/checkpoints  (a stand-in name: pick your own file)

P1 - Load Image and VAE Encode.json
    model   waiIllustriousSDXL_v170.safetensors  ->  models/checkpoints
    input   course_p1_cat.png

P1 - Noise into a picture.json
    model   waiIllustriousSDXL_v170.safetensors  ->  models/checkpoints

P1 - Save vs Preview.json
    model   waiIllustriousSDXL_v170.safetensors  ->  models/checkpoints

P1 - Starter (seven nodes).json
    model   waiIllustriousSDXL_v170.safetensors  ->  models/checkpoints

Model downloads
    Every model file the workflows above use: the folder it goes in, and its download link. Put the file in
    that folder, then press R in ComfyUI. The links are inside the workflows too, so newer versions of ComfyUI
    offer a download button for a file that is missing when you open the workflow.

z_image_turbo_aio.safetensors  ->  models/checkpoints   (10.3 GB, community repack, not official)
    https://huggingface.co/SeeSee21/Z-Image-Turbo-AIO/resolve/main/z-image-turbo-fp8-aio.safetensors
    Save it as z_image_turbo_aio.safetensors. Z-Image Turbo has no official one-file version: its makers
    and Comfy-Org publish it as three files (Comfy-Org/z_image_turbo on Hugging Face), which ComfyUI's own
    template "Z-Image-Turbo: Text to Image" downloads for you. This one-file repack is by SeeSee21 (Apache 2.0).

waiIllustriousSDXL_v170.safetensors  ->  models/checkpoints   (6.9 GB, Civitai, WAI-illustrious-SDXL v17.0 by WAI0731)
    https://civitai.com/api/download/models/2883731

qwen_image_2.1_int8_convrot.safetensors  ->  models/diffusion_models   (7.3 GB, Comfy-Org)
    https://huggingface.co/Comfy-Org/Qwen-Image-2.1/resolve/main/diffusion_models/qwen_image_2.1_int8_convrot.safetensors

qwen3vl_8b_int8_convrot.safetensors  ->  models/text_encoders   (9.4 GB, Comfy-Org)
    https://huggingface.co/Comfy-Org/Qwen-Image-2.1/resolve/main/text_encoders/qwen3vl_8b_int8_convrot.safetensors

qwen_image_2.1_vae_bf16.safetensors  ->  models/vae   (676 MB, Comfy-Org)
    https://huggingface.co/Comfy-Org/Qwen-Image-2.1/resolve/main/vae/qwen_image_2.1_vae_bf16.safetensors

Installing on a Mac, an AMD card or a cloud GPU: the written guides (the "Written guide" cards in Modules 2 and 3)
    Mac          https://stuubzzz.studio/comfyui-install/#mac
    AMD          https://stuubzzz.studio/comfyui-install/#amd
    Cloud GPU    https://stuubzzz.studio/comfyui-install/#cloud

Parts 2 and 3 go further. The full course: https://stuubzzz.studio/comfyui-course/
    (It is also included in the Studio tier on Patreon: patreon.com/Stuubzzz)
