MCPcopy Create free account

hub / github.com/Lightricks/ComfyUI-LTXVideo / functions

Functions429 in github.com/Lightricks/ComfyUI-LTXVideo

⨍ Functions 429◇ Types & classes 117
↓ 1 callersFunction_render_resolution
Compute the higher render resolution that preserves aspect ratio.
sparse_tracks.py:189
↓ 1 callersMethod_reshape_masks
Reshapes input masks to consistent 4D format.
latents.py:296
↓ 1 callersMethod_resize_masks
Resizes all masks to match latent dimensions.
latents.py:313
↓ 1 callersFunction_resize_preserving_aspect_ratio
( images: torch.Tensor, long_side: int, mode: str )
pyramid_blending.py:48
↓ 1 callersMethod_save_exr_frames
( hdr_image: torch.Tensor, output_dir: str, filename_prefix: str, half_precisi
hdr.py:201
↓ 1 callersFunction_set_guide_attention_entries
Write guide_attention_entries into conditioning (immutable update).
iclora_attention.py:13
↓ 1 callersMethod_validate_dimensions
Validates that latent dimensions match except for frames.
latents.py:165
↓ 1 callersFunction_wrap_define_schema
Wrap the define_schema method to inject the display_name into the schema.
nodes_registry.py:47
↓ 1 callersFunctionapplyLabels
()
web/js/guider_parameters_advanced.js:68
↓ 1 callersFunctionbuildContextMenu
()
web/js/sparse_track_editor.js:337
↓ 1 callersMethodcalc_stg_indexes
( self, run_vx: bool, run_ax: bool, audio_ptb: bool, video_ptb: bool,
guiders/multimodal_guider.py:63
↓ 1 callersFunctioncalculate_temporal_output_boundaries
Calculate temporal output boundaries for the decoded tile. Args: overlap_start: Starting frame index including overlap time_scale
tiled_vae_decode.py:249
↓ 1 callersFunctioncatmullRom
(p0, p1, p2, p3, t)
web/js/sparse_track_editor.js:7
↓ 1 callersFunctioncheck_deprecated
()
q8_nodes.py:38
↓ 1 callersFunctioncheck_q8_available
()
q8_nodes.py:30
↓ 1 callersFunctionclean_response
(text)
gemma_encoder.py:283
↓ 1 callersFunctionclosestSegment
(pts, p)
web/js/sparse_track_editor.js:85
↓ 1 callersFunctioncollectModalityParams
(node)
web/js/guider_parameters_advanced.js:188
↓ 1 callersFunctioncompute_chunk_boundaries
Compute chunk boundaries for temporal tiling. Args: chunk_start: Starting frame index for the current chunk temporal_tile_length:
tiled_vae_decode.py:214
↓ 1 callersMethodcreate_embeddings
(self, features, attention_mask)
text_embeddings_connectors.py:246
↓ 1 callersMethoddecompress
(self, z: Tensor)
hdr.py:48
↓ 1 callersFunctiondecorator
(node_class: Type)
nodes_registry.py:92
↓ 1 callersMethoddo_modality
(self)
guiders/parameters.py:64
↓ 1 callersMethoddo_perturbed
(self)
guiders/parameters.py:61
↓ 1 callersMethoddo_uncond
(self)
guiders/parameters.py:58
↓ 1 callersMethoddown_load_image_captioner
(self, image_captioner, load_device)
prompt_enhancer_nodes.py:129
↓ 1 callersMethoddown_load_llm_model
(self, llm_name, load_device)
prompt_enhancer_nodes.py:116
↓ 1 callersMethodencode_image
(self, image, shape, vae)
latents.py:536
↓ 1 callersFunctionenhance_i2v
Enhance a text prompt for I2V generation using a reference image.
gemma_encoder.py:469
↓ 1 callersFunctionenhance_t2v
Enhance a text prompt for T2V generation.
gemma_encoder.py:451
↓ 1 callersFunctionextract_model_id
(ckpt_name: str)
gemma_api_conditioning.py:29
↓ 1 callersMethodfind_step
(self, sigma: torch.Tensor, step_sigmas: torch.Tensor)
dynamic_conditioning.py:30
↓ 1 callersFunctiongaussian_blur_2d
(img, kernel_size, sigma)
tricks/nodes/ltx_pag_node.py:13
↓ 1 callersFunctiongenerate_cinematic_prompt
( image_caption_model, image_caption_processor, prompt_enhancer_model, prompt_enhancer_tokeniz
prompt_enhancer_utils.py:75
↓ 1 callersMethodgenerate_freq_grid
(self, spacing, dtype, device)
embeddings_connector.py:219
↓ 1 callersFunctionget_alphacumprod
(sigma)
tricks/utils/noise_utils.py:1
↓ 1 callersFunctionget_feta_scores
(img_q, img_k, num_heads, transformer_options)
tricks/utils/feta_enhance_utils.py:32
↓ 1 callersFunctionget_flowedit_sample
(skip_steps, refine_steps, seed)
tricks/nodes/ltx_flowedit_nodes.py:81
↓ 1 callersMethodget_fractional_positions
(self, indices_grid)
embeddings_connector.py:176
↓ 1 callersFunctionget_sample_forward
( gamma, start_step, end_step, gamma_trend, seed, attn_bank=None, order="first" )
tricks/nodes/rectified_sampler_nodes.py:24
↓ 1 callersFunctionget_sample_forward
(attn_bank, save_steps, single_layers, double_layers)
tricks/nodes/rf_edit_sampler_nodes.py:6
↓ 1 callersFunctionget_sample_reverse
( latent_image, eta, start_time, end_time, eta_trend, attn_bank=None, order="first" )
tricks/nodes/rectified_sampler_nodes.py:98
↓ 1 callersFunctionget_sample_reverse
(attn_bank, inject_steps, single_layers, double_layers)
tricks/nodes/rf_edit_sampler_nodes.py:77
↓ 1 callersMethodget_transformer_blocks
(model: ModelPatcher)
stg.py:217
↓ 1 callersMethodget_transformer_blocks
(model: ModelPatcher)
guiders/multimodal_guider.py:53
↓ 1 callersFunctionget_video_latent_blend_coefficients
Returns a blend coefficient list shaped such that: - It is 0.0 outside the range [video_frame_index_start, video_frame_index_end] - It ra
latents.py:578
↓ 1 callersFunctionimportXlsxValues
(node, filename)
web/js/guider_parameters_advanced.js:124
↓ 1 callersFunctioninitEditor
(node)
web/js/sparse_track_editor.js:155
↓ 1 callersFunctioninjectMultilineLabels
(node)
web/js/guider_parameters_advanced.js:59
↓ 1 callersFunctioninject_model
(diffusion_model)
tricks/modules/ltx_model.py:250
↓ 1 callersFunctionis_integer
(string)
tricks/nodes/attn_override_node.py:1
↓ 1 callersFunctionlist_in_name
(check_list, name)
q8_nodes.py:26
↓ 1 callersFunctionloadBgImage
(node, b64)
web/js/sparse_track_editor.js:262
↓ 1 callersFunctionload_audio_embeddings_connector
(sd, transformer_config, dtype=torch.bfloat16)
embeddings_connector.py:420
↓ 1 callersMethodload_lora
(self, model, lora_name, strength_model)
q8_nodes.py:168
↓ 1 callersMethodload_model
(self, gemma_path: str, ltxv_path: str, max_length: int)
gemma_encoder.py:244
↓ 1 callersFunctionload_stg_presets
()
stg.py:541
↓ 1 callersFunctionload_text_embeddings_pipeline
Load feature extractor + embeddings processor from LTX-V checkpoint. Auto-detects model variant (19B/22B, video-only/AV). Args: ltxv
text_embeddings_connectors.py:316
↓ 1 callersFunctionload_video_embeddings_connector
(sd, transformer_config, dtype=torch.bfloat16)
embeddings_connector.py:383
↓ 1 callersFunctionloop
()
web/js/sparse_track_editor.js:592
↓ 1 callersFunctionltxv_gemma_clip
(encoder_path, ltxv_path, processor=None, dtype=None)
gemma_encoder.py:171
↓ 1 callersMethodmemory_required
(self, input_shape)
prompt_enhancer_nodes.py:60
↓ 1 callersFunctionnorm_and_concat_padded_batch
Normalize a 4D tensor [B, T, D, L] per sample and per layer, using sequence_lengths to mask. Returns [B, T, D * L] tensor with original padd
text_embeddings_connectors.py:75
↓ 1 callersFunctionnorm_and_concat_per_token_rms
Per-token RMSNorm normalization. Each token is normalized independently using RMSNorm over its D dimension. This is naturally causal and sup
text_embeddings_connectors.py:133
↓ 1 callersMethodoptional
(s)
latent_norm.py:220
↓ 1 callersMethodparse_stg_layers_indices
(cls, stg_layers_indices: str)
stg.py:681
↓ 1 callersMethodpatch_model
(cls, model: ModelPatcher, stg_flag: STGFlag)
guiders/multimodal_guider.py:44
↓ 1 callersFunctionpopulateWidgets
(node, data)
web/js/guider_parameters_advanced.js:111
↓ 1 callersMethodprecompute_freqs
(self, indices_grid, spacing)
embeddings_connector.py:186
↓ 1 callersMethodprecompute_freqs_cis
(self, indices_grid, spacing="exp")
embeddings_connector.py:246
↓ 1 callersMethodprocess
(self, samples1, samples2, overlap, axis=0)
easy_samplers.py:871
↓ 1 callersFunctionproject
(v0: torch.Tensor, v1: torch.Tensor)
stg.py:46
↓ 1 callersFunctionregister_node
Register a ComfyUI node class to ComfyUI's global nodes' registry. Args: node_class (Type): The class of the node to be registered.
nodes_registry.py:14
↓ 1 callersFunctionreloadState
(node)
web/js/sparse_track_editor.js:243
↓ 1 callersFunctionrender
(node)
web/js/sparse_track_editor.js:624
↓ 1 callersFunctionresizeCanvas
(node)
web/js/sparse_track_editor.js:603
↓ 1 callersMethodset_cfgs
(self, **kwargs)
tricks/nodes/ltx_flowedit_nodes.py:14
↓ 1 callersMethodset_conds
(self, positive, negative)
stg.py:374
↓ 1 callersMethodset_conds
(self, positive, negative)
stg.py:815
↓ 1 callersMethodset_conds
(self, **kwargs)
tricks/nodes/ltx_flowedit_nodes.py:11
↓ 1 callersMethodset_conds
(self, positive, negative)
guiders/multimodal_guider.py:60
↓ 1 callersFunctionsetupEvents
(node)
web/js/sparse_track_editor.js:473
↓ 1 callersFunctionsetupExportButton
(node)
web/js/guider_parameters_advanced.js:223
↓ 1 callersFunctionsetupUploadButton
(node)
web/js/guider_parameters_advanced.js:139
↓ 1 callersFunctionshowMenu
(node, canvasEvt, imgPos)
web/js/sparse_track_editor.js:355
↓ 1 callersMethodsigma_to_params_mapping
(self, sigma)
stg.py:359
↓ 1 callersFunctionstartRenderLoop
(node)
web/js/sparse_track_editor.js:589
↓ 1 callersMethodstatistical_normalize
( self, latents, target_mean, target_std, percentile, factor, clip_outliers )
latent_norm.py:134
↓ 1 callersFunctiontensor_to_pil
(tensor)
prompt_enhancer_utils.py:58
↓ 1 callersFunctiontensor_to_pil
Convert ComfyUI image tensor to PIL Image.
gemma_encoder.py:43
↓ 1 callersFunctiontransformers_gemma3_from_encoder
(encoder)
gemma_encoder.py:588
MethodINPUT_TYPES
(s)
low_vram_loaders.py:31
MethodINPUT_TYPES
(cls)
gemma_api_conditioning.py:61
MethodINPUT_TYPES
(s)
latents.py:26
MethodINPUT_TYPES
(s)
latents.py:106
MethodINPUT_TYPES
(s)
latents.py:226
MethodINPUT_TYPES
(s)
latents.py:337
MethodINPUT_TYPES
(s)
latents.py:403
MethodINPUT_TYPES
(s)
latents.py:511
MethodINPUT_TYPES
(cls)
latents.py:638
← previousnext →101–200 of 429, ranked by callers