WorkflowListedExternal

LTXV 13b distilled 0.9.7 fp8 img2video improved workflow

I was getting terrible results with the basic workflow for LTXV distilled fp8, so I added florence caption and image resize.

TaskImage to imageFamilyLTX-VideoEstimated VRAMEstimate unavailableBgrade

Runtime profile

Source excerpt

I was getting terrible results with the basic workflow for LTXV distilled fp8, so I added florence caption and image resize.

Estimated VRAM requirement

Estimate unavailable

9.12 GB across 1 of 3 model files. Model file total + 25% loading overhead + 2 GB execution buffer, rounded up.

Requirements

11 requirements

AI-generated commentary

AI-generated explanation based on source and configuration details. Suggestions are clearly labeled.

Use this workflow for image-to-video work: the “img2video” label indicates an image input and video output, while the description says it adds florence caption and image resize to the basic LTXV distilled fp8 workflow.
Suggestion · not verified

Verify that the workflow accepts an image and returns video; this behavior is untested.
Suggestion · not verified

Find it on Civitai at https://civitai.com/models/1581664; the listed publisher is DjSaKaS.

The listed model files are ltxv-13b-0.9.7-distilled-fp8.safetensors, ltxv-spatial-upscaler-0.9.7.safetensors, and t5xxl_fp16.safetensors.
Listed node packs include ComfyUI-LTXVideo, comfyui-videohelpersuite, comfyui_layerstyle, and comfyui-kjnodes.
Other listed node pack names are comfyui-custom-scripts, image-saver, pr-was-node-suite-comfyui-47064894, and was-node-suite-comfyui.

Before use, verify that ltxv-13b-0.9.7-distilled-fp8.safetensors and ltxv-spatial-upscaler-0.9.7.safetensors are available in your setup.
Suggestion · not verified
Before use, verify that each listed node pack is available in your setup.
Suggestion · not verified

Does this need to be edited?

Sources

1 source

CivitaiSource

Verified Jul 19

Source excerpts

1 excerpt

Source context: 1056 downloads · Type Workflows · Base model LTXV