Nodes/comfui-helto-ltx/LTX 2.3 Image Guides (All-in-One)
ComfyUI Node

LTX 2.3 Image Guides (All-in-One)

All-in-one native LTXV guide node: select images, apply guide metadata, and output guided conditioning plus latent.

By helto4real·Created 3 months ago·Updated 2 months ago· 0
LTX 2.3 Image Guides (All-in-One)
  • positive
  • negative
  • vae
  • latent
  • start_images
  • positive
  • negative
  • latent
width768
height512
half_size_first_passfalse
fps24.00
num_frames97
timing_modeframe
resize_modecontain
duplicate_policyerror
pad_color0,0,0
img_compression35
global_strength1.00
lock_start_framesfalse
lock_end_framefalse
start_images_strength0.85
guides_json{"version":1,"guides":[]}
CategoryLTX 2.3

Inputs (20)

NameTypeDefaultDescription
positiveCONDITIONINGPositive conditioning to augment with native LTXV guide metadata.
negativeCONDITIONINGNegative conditioning to augment with matching native LTXV guide metadata.
vaeVAELTXV VAE used to encode selected guide images and optional start image sequences.
widthINT76864–16384Target output size for this stage. With half_size_first_pass enabled and no latent connected, the internal latent uses half this value.
heightINT51264–16384Target output size for this stage. With half_size_first_pass enabled and no latent connected, the internal latent uses half this value.
half_size_first_passBOOLEANfalseFor 2x LTX upscale workflows: when no latent is connected, create and guide a half-size first-pass latent. Width/height should be the final target size.
fpsFLOAT24.001–240Frames per second used when timing_mode is seconds.
num_framesINT971–16384Pixel frame count used for timing, negative frame positions, and internally-created empty latents. Native LTXV lengths must be 8*n + 1, for example 97, 105, 113.
timing_modeCOMBOframeInterpret manual guide positions as frame indexes or seconds.
resize_modeCOMBOcontainHow guide images are resized before VAE encoding. contain/pad preserves aspect ratio with padding.
duplicate_policyCOMBOerrorHow to handle manual guide images that resolve to the same frame.
pad_colorSTRING0,0,0RGB padding color for contain/pad resize mode. Accepts r,g,b or #rrggbb.
img_compressionINT350–100Native LTXV image compression applied before guide encoding. Set 0 to disable.
global_strengthFLOAT1.000–1Multiplier applied to every manual guide strength and start sequence strength.
lock_start_framesBOOLEANfalseWhen enabled, frame 0 guides and start_images are written into the beginning video latent instead of only appended as guide references. VAE-level lock, not pixel-perfect copy.
lock_end_frameBOOLEANfalseWhen enabled, a manual guide resolving to the final frame is written into the final video latent instead of only appended as a guide reference. VAE-level lock, not pixel-perfect copy.
start_images_strengthFLOAT0.850–1Strength for the optional start image sequence before global_strength is applied.
guides_jsonSTRING{"version":1,"guides":[]}Hidden serialized guide data used by the custom UI and saved in workflows.
latentoptLATENTOptional existing video latent. When connected, its shape is used and half_size_first_pass does not resize it.
start_imagesoptIMAGEOptional IMAGE batch from a video source. Applied as a native multi-frame guide starting at frame 0.

Outputs (3)

NameTypeDescription
positiveCONDITIONING
negativeCONDITIONING
latentLATENT