LTXVContextWindows - ComfyUI Built-in Node Documentation

This node sets context windows for LTXV-like models during sampling.

This node sets context windows for LTXV-like models during sampling. It divides generation into overlapping windows to help manage memory usage and improve temporal coherence.

Inputs

Parameter	Description	Data Type	Required	Range
model	The model to apply context windows to during sampling.	MODEL	Yes	-
context_length	The length of the context window in real frames. Must be 8*n + 1. (default: 145)	INT	Yes	Minimum: 1 Maximum: nodes.MAX_RESOLUTION Step: 8
context_overlap	The overlap of the context window in real frames. (default: 40)	INT	Yes	Minimum: 0 Step: 8
context_schedule	Step-dependent scheduling algorithm for context windows. (default: “UNIFORM_STANDARD”)	COMBO	Yes	"STATIC_STANDARD" "UNIFORM_STANDARD" "UNIFORM_LOOPED" "BATCHED"
context_stride	The stride of the context window; only applicable to uniform schedules. (default: 1)	INT	Yes	Minimum: 1
closed_loop	Whether to close the context window loop; only applicable to looped schedules. (default: False)	BOOLEAN	Yes	True False
fuse_method	The method to use to fuse the context windows. Available options are defined by ContextFuseMethods.LIST_STATIC. (default: “PYRAMID”)	COMBO	Yes	Defined by ContextFuseMethods.LIST_STATIC
freenoise	Whether to apply FreeNoise noise shuffling, improves window blending. (default: True)	BOOLEAN	Yes	True False
retain_first_frame	Retain the first latent frame in every context window (may help retain initial reference). (default: False)	BOOLEAN	Yes	True False
split_conds_to_windows	Whether to split multiple conditionings (created by ConditionCombine) to each window based on region index. (default: False)	BOOLEAN	Yes	True False

Note: The context_length value is provided in real frames and is internally converted to latent frames using the formula ((context_length - 1) // 8) + 1, with a minimum of 1. The context_overlap value is also provided in real frames and is converted to latent frames using integer division by 8, with a minimum of 0. The tooltip for context_length indicates it must follow the pattern 8*n + 1.

Outputs

Output Name	Description	Data Type
MODEL	The model with context windows applied for sampling.	MODEL

This documentation was AI-generated. If you find any errors or have suggestions for improvement, please feel free to contribute! Edit on GitHub

---

Source fingerprint (SHA-256): 148649d0a938e08c932a163f5d7614332626fba37b8f79db7f92bbcf422e692f