SquiDavey's ZIT Workflow (Local LLM / Styles / T2I / SeedVR2 / Post-Process) — v1.0

This ComfyUI workflow is for creating images from a text concept or processing an existing image: it can expand prompts with a local LLM, generate with Z-Image Turbo, add CRT/VHS effects, and upscale with SeedVR; it takes a text concept or uploaded image and outputs a generated,

For a new image, LM Studio expands a simple concept into a rich, photorealistic description, Z-Image Turbo creates the base image, CRT/VHS processing adds analog effects, and SeedVR performs a tiled upscale.

For an existing image, upload it to the UPSCALE BYPASS node and use the image-to-upscale route.

The text-to-image route accepts a simple concept in the Prompt group and passes it to LM Studio for expansion.

The image-to-upscale route accepts an uploaded image in UPSCALE BYPASS and applies CRT/VHS simulation and tiled SeedVR upscaling.

For text-to-image use, download a model such as Qwen 2.5 7B in LM Studio, open Developer/Server, and start the server on port 1234.

Place z_image_turbo_bf16.safetensors in models/diffusion_models/, ae.safetensors and ema_vae_fp16.safetensors in models/vae/, qwen_3_4b.safetensors in models/text_encoders/, seedvr2_ema_7b_fp16.safetensors in models/checkpoints/, and 4x-ClearRealityV1.pth in models/upscale_models

The workflow calls for these model files: ae.safetensors, ema_vae_fp16.safetensors, qwen_3_4b.safetensors, seedvr2_ema_7b_fp16.safetensors, and z_image_turbo_bf16.safetensors.

It also lists these node packs: comfyui_essentials, comfyui_ttp_toolset, comfyui-crystools, comfyui-custom-scripts, comfyui-easy-use, comfyui-kjnodes, comfyui-lmstudio-image-to-text-node, comfyui-mxtoolkit, crt-nodes, rgthree-comfy, and seedvr2_videoupscaler.

For new generation, set the SEEDVR / LOAD IMAGE switch to True, enter a concept in the Prompt group, and press Queue Prompt.

For existing-image upscaling, set the switch to False, upload your image to UPSCALE BYPASS, and optionally disable the LLM group.

If upscaling runs out of VRAM, try reducing the SeedVR tile size and increasing the blocks to swap.

Suggestion · non vérifié

Before using SeedVR, verify that ema_vae_fp16.safetensors and seedvr2_ema_7b_fp16.safetensors are the files you intend to use.

Suggestion · non vérifié

The listing does not specify a license for this workflow.