ComfyUI Node: VRGDG LTX Audio Video LoRA Train Chunk

Authored by vrgamegirl19

Created

Updated

523 stars

Run ComfyUI workflows without the setup

No installs, no CUDA version roulette, no GPU sitting idle on your bill. Bring a workflow and run it in the browser.

Category

VRGDG/Training

Inputs

model MODEL
dataset_videos_dir STRING
workspace_dir STRING
run_name STRING
output_name STRING
av_profile
  • fast_short_clip
  • step200
  • balanced
  • full_video
  • overnight_full_video
use_profile_defaults BOOLEAN
resolution_width INT
resolution_height INT
target_frames STRING
frame_extraction
  • head
  • chunk
  • slide
  • uniform
  • full
max_frames INT
target_fps FLOAT
steps_per_run INT
total_target_steps INT
network_dim INT
network_alpha INT
blocks_to_swap INT
separate_audio_buckets BOOLEAN
clear_memory_before_gemma BOOLEAN
lora_target_preset
  • t2v
  • v2v
  • audio
  • full
fp8_base BOOLEAN
fp8_scaled BOOLEAN
learning_rate_preset
  • Custom
  • 1e-4
  • 7e-5
  • 5e-5
  • 3e-5
  • 1e-5
learning_rate FLOAT
num_repeats INT
cache_strategy
  • auto
  • force
  • skip
copy_latest_to_comfy_loras BOOLEAN
keep_only_comfy_lora BOOLEAN
strength_model FLOAT
create_captions BOOLEAN
caption_text STRING
add_trigger_word BOOLEAN
trigger_text STRING
musubi_root STRING
ltx2_checkpoint STRING
gemma_root STRING
gemma_load_in_4bit BOOLEAN
gemma_recovery_mode BOOLEAN

Outputs

MODEL

STRING

STRING

STRING

STRING

INT

INT

Extension: VRGameDevGirl Video Enhancement Nodes

Film grain and color match nodes designed for high-quality frame-by-frame video enhancement in ComfyUI.

Authored by vrgamegirl19

Looking for a different node?

Run ComfyUI workflows without the setup

No installs, no CUDA version roulette, no GPU sitting idle on your bill. Bring a workflow and run it in the browser.

Learn more