1019703 — v1.0

EML_LTX_TTV/ITV is a ComfyUI workflow for creating LTX-Video videos from text prompts or uploaded images, with Text To Video and Image To Video modes and WebP or MP4 output.

In image mode, Florence2 supplies a static image description, while Llama 3.2 turns your prompt into movement and lighting cues.

For generated images, the guide says CRF adds noise to help address frozen video.

The guide lists prompts and uploaded images as inputs; it says videos and temporary PNG images are saved in `ComfyUI\output`.

If custom nodes are missing at startup, the guide recommends using Manager → Install Missing Custom Nodes, selecting all, installing them, and restarting when prompted.

Suggestion · not verified

For Llama 3.2 prompt generation, the guide recommends installing Ollama, running `ollama run llama3.2`, and activating `ltxv` mode.

Suggestion · not verified

The workflow archive lists the model files `ltx-2b-v0.9-bf16.safetensors`, `microsoft/Florence-2-large-ft`, and `t5xxl_fp8_e4m3fn.safetensors`.

The guide lists defaults of 768×512 at 3:2, 25 fps, 97 frames, 30 steps, and CFG 3.0; the aspect-ratio list includes a 16:9 hint.

For adjustments, the guide points to groups marked in blue and advises changing only `steps` in `LTXVScheduler`.

Suggestion · not verified

The guide points to `LTXV Model Configurator` as a reference for resolution choices and gives CFG 3.0–3.5 as a recommended range.

Manual prompt mode is in development; the guide says the workflow currently supports LLM input.

The guide advertises use with 8 GB VRAM, but warns that this workflow needs more video memory and is lower quality than the manual workflow.