Nodes/ComfyUI_agilly1989_motorway/LTXVConditioning_motorway_edition
ComfyUI Node

LTXVConditioning_motorway_edition

Tell LTXV how fast your video should feel — frame rate on conditioning

By agilly1989·Created 2 years ago·Updated about a year ago· 7
LTXVConditioning_motorway_edition
  • MOTORWAY 🚌💨
  • MOTORWAY 🚌💨
INPUT_positive_keypositive
INPUT_negative_keynegative
frame_rate25.00
OUTPUT_positive_keypositive
OUTPUT_negative_keynegative

LTX Video is the speed tier of local video generation - where Wan takes minutes per clip, LTX drafts in seconds, which is exactly why people use it for quick motion tests before refining elsewhere. And LTXVConditioning_motorway_edition is the small but important step that happens between "I have my prompt conditioning" and "let's sample": it stamps a frame rate onto your positive and negative conditioning.

Why does that matter? Video models bake temporal behavior into their conditioning, and LTXV is no exception - the frame rate is a first-class value that the sampler uses to shape how motion develops over time. Set it to match the speed you want the finished clip to read as. The default is 25 fps, which is the conventional video standard; drop toward 12-ish for a more stuttery, deliberate look or raise it for smoother motion. In the core implementation this is literally a conditioning_set_values call that tags both positive and negative with the same frame_rate, so it's a cheap node - but skip it and your LTXV workflow is running without a piece of its identity.

The wrapper reads both conditioning vectors out of your Motorway pipe by key, stamps them, and writes them back. Same values in, same values out, plus the frame rate.

The wrapper

  • MOTORWAY 🚌💨 - the pipe. Required.
  • INPUT_positive_key (default positive) - where the positive conditioning lives.
  • INPUT_negative_key (default negative) - where the negative conditioning lives.
  • frame_rate (FLOAT, 0–1000, default 25) - the fps to stamp.
  • OUTPUT_positive_key (default positive) and OUTPUT_negative_key (default negative) - where the tagged conditioning lands.

Outputs stay at the same keys by default, so the node can slot into a chain and the next node picks up the updated conditioning without any re-keying. To hand the result to a normal node, pull positive/negative out with a Motorway AxB ramp.

Installing

Part of the Motorway pack:

cd ComfyUI/custom_nodes
git clone https://github.com/agilly1989/ComfyUI_agilly1989_motorway
# restart ComfyUI

Or ComfyUI Manager → search ComfyUI_agilly1989_motorway → Install. No model downloads for the node - but LTX itself needs its checkpoints, and you'll also want the LTXV model-sampling setup from core before sampling anything.

Gotchas

  • The conditioning must already be in the pipe - Key: 'positive' doesn't exist in motorway means your CLIP/text-encode step never wrote it.
  • Keys are exact strings; the defaults write back to the same keys, so most chains just work.
  • This node only does the conditioning tag. It doesn't encode video, doesn't set up the latent, doesn't sample. Pair it with LTXVImgToVideo_motorway_edition (or the text-to-video latent node) and a KSampler for a full pipeline.
  • Beta pack: author's README says "IF THINGS BREAK ITS BECAUSE I BROKE IT," and generated clones have been toggled off in some releases. Missing node after install = beta, not you.

Frame rate is one of those settings that reads "minor" until you see two otherwise-identical clips at different fps. This node is how you own that choice inside the pipe.

Categoryagilly1989 Nodes/Motorway-ed/conditioning/video_models

Inputs (6)

NameTypeDefaultDescription
MOTORWAY 🚌💨MOTORWAY 🚌💨
INPUT_positive_keySTRINGpositive
INPUT_negative_keySTRINGnegative
frame_rateFLOAT25.000–1000
OUTPUT_positive_keySTRINGpositive
OUTPUT_negative_keySTRINGnegative

Outputs (1)

NameTypeDescription
MOTORWAY 🚌💨MOTORWAY 🚌💨