theluyuan/DiffSynth-Studio
1
0
Fork 0
mirror of https://github.com/modelscope/DiffSynth-Studio.git synced 2026-03-18 22:08:13 +00:00
Code Issues Packages Projects Releases Wiki Activity

Merge branch 'main' into layercontrol_v2

Browse Source
This commit is contained in:
Zhongjie Duan
2026-03-03 21:04:04 +08:00
committed by GitHub
parent 07f5d88ac9 ab8f455c46
commit 6bcb99fd2e
81 changed files with 4118 additions and 124 deletions
Download Patch File Download Diff File Expand all files Collapse all files

19
examples/anima/model_inference/anima-preview.py Normal file
View File

@@ -0,0 +1,19 @@
from diffsynth.pipelines.anima_image import AnimaImagePipeline, ModelConfig
import torch
pipe = AnimaImagePipeline.from_pretrained(
torch_dtype=torch.bfloat16,
device="cuda",
model_configs=[
ModelConfig(model_id="circlestone-labs/Anima", origin_file_pattern="split_files/diffusion_models/anima-preview.safetensors"),
ModelConfig(model_id="circlestone-labs/Anima", origin_file_pattern="split_files/text_encoders/qwen_3_06b_base.safetensors"),
ModelConfig(model_id="circlestone-labs/Anima", origin_file_pattern="split_files/vae/qwen_image_vae.safetensors"),
],
tokenizer_config=ModelConfig(model_id="Qwen/Qwen3-0.6B", origin_file_pattern="./"),
tokenizer_t5xxl_config=ModelConfig(model_id="stabilityai/stable-diffusion-3.5-large", origin_file_pattern="tokenizer_3/")
)
prompt = "Masterpiece, best quality, solo, long hair, wavy hair, silver hair, blue eyes, blue dress, medium breasts, dress, underwater, air bubble, floating hair, refraction, portrait."
negative_prompt = "worst quality, low quality, monochrome, zombie, interlocked fingers, Aissist, cleavage, nsfw,"
image = pipe(prompt, seed=0, num_inference_steps=50)
image.save("image.jpg")

30
examples/anima/model_inference_low_vram/anima-preview.py Normal file
View File

@@ -0,0 +1,30 @@
from diffsynth.pipelines.anima_image import AnimaImagePipeline, ModelConfig
import torch
vram_config = {
"offload_dtype": "disk",
"offload_device": "disk",
"onload_dtype": "disk",
"onload_device": "disk",
"preparing_dtype": torch.bfloat16,
"preparing_device": "cuda",
"computation_dtype": torch.bfloat16,
"computation_device": "cuda",
}
pipe = AnimaImagePipeline.from_pretrained(
torch_dtype=torch.bfloat16,
device="cuda",
model_configs=[
ModelConfig(model_id="circlestone-labs/Anima", origin_file_pattern="split_files/diffusion_models/anima-preview.safetensors", **vram_config),
ModelConfig(model_id="circlestone-labs/Anima", origin_file_pattern="split_files/text_encoders/qwen_3_06b_base.safetensors", **vram_config),
ModelConfig(model_id="circlestone-labs/Anima", origin_file_pattern="split_files/vae/qwen_image_vae.safetensors", **vram_config),
],
tokenizer_config=ModelConfig(model_id="Qwen/Qwen3-0.6B", origin_file_pattern="./"),
tokenizer_t5xxl_config=ModelConfig(model_id="stabilityai/stable-diffusion-3.5-large", origin_file_pattern="tokenizer_3/"),
vram_limit=torch.cuda.mem_get_info("cuda")[1] / (1024 ** 3) - 0.5,
)
prompt = "Masterpiece, best quality, solo, long hair, wavy hair, silver hair, blue eyes, blue dress, medium breasts, dress, underwater, air bubble, floating hair, refraction, portrait."
negative_prompt = "worst quality, low quality, monochrome, zombie, interlocked fingers, Aissist, cleavage, nsfw,"
image = pipe(prompt, seed=0, num_inference_steps=50)
image.save("image.jpg")

14
examples/anima/model_training/full/anima-preview.sh Normal file
View File

@@ -0,0 +1,14 @@
accelerate launch examples/anima/model_training/train.py \
--dataset_base_path data/example_image_dataset \
--dataset_metadata_path data/example_image_dataset/metadata.csv \
--max_pixels 1048576 \
--dataset_repeat 50 \
--model_id_with_origin_paths "circlestone-labs/Anima:split_files/diffusion_models/anima-preview.safetensors,circlestone-labs/Anima:split_files/text_encoders/qwen_3_06b_base.safetensors,circlestone-labs/Anima:split_files/vae/qwen_image_vae.safetensors" \
--tokenizer_path "Qwen/Qwen3-0.6B:./" \
--tokenizer_t5xxl_path "stabilityai/stable-diffusion-3.5-large:tokenizer_3/" \
--learning_rate 1e-5 \
--num_epochs 2 \
--remove_prefix_in_ckpt "pipe.dit." \
--output_path "./models/train/anima-preview_full" \
--trainable_models "dit" \
--use_gradient_checkpointing

16
examples/anima/model_training/lora/anima-preview.sh Normal file
View File

@@ -0,0 +1,16 @@
accelerate launch examples/anima/model_training/train.py \
--dataset_base_path data/example_image_dataset \
--dataset_metadata_path data/example_image_dataset/metadata.csv \
--max_pixels 1048576 \
--dataset_repeat 50 \
--model_id_with_origin_paths "circlestone-labs/Anima:split_files/diffusion_models/anima-preview.safetensors,circlestone-labs/Anima:split_files/text_encoders/qwen_3_06b_base.safetensors,circlestone-labs/Anima:split_files/vae/qwen_image_vae.safetensors" \
--tokenizer_path "Qwen/Qwen3-0.6B:./" \
--tokenizer_t5xxl_path "stabilityai/stable-diffusion-3.5-large:tokenizer_3/" \
--learning_rate 1e-4 \
--num_epochs 5 \
--remove_prefix_in_ckpt "pipe.dit." \
--output_path "./models/train/anima-preview_lora" \
--lora_base_model "dit" \
--lora_target_modules "" \
--lora_rank 32 \
--use_gradient_checkpointing

145
examples/anima/model_training/train.py Normal file
View File

@@ -0,0 +1,145 @@
import torch, os, argparse, accelerate
from diffsynth.core import UnifiedDataset
from diffsynth.pipelines.anima_image import AnimaImagePipeline, ModelConfig
from diffsynth.diffusion import *
os.environ["TOKENIZERS_PARALLELISM"] = "false"
class AnimaTrainingModule(DiffusionTrainingModule):
def __init__(
self,
model_paths=None, model_id_with_origin_paths=None,
tokenizer_path=None, tokenizer_t5xxl_path=None,
trainable_models=None,
lora_base_model=None, lora_target_modules="", lora_rank=32, lora_checkpoint=None,
preset_lora_path=None, preset_lora_model=None,
use_gradient_checkpointing=True,
use_gradient_checkpointing_offload=False,
extra_inputs=None,
fp8_models=None,
offload_models=None,
device="cpu",
task="sft",
):
super().__init__()
# Load models
model_configs = self.parse_model_configs(model_paths, model_id_with_origin_paths, fp8_models=fp8_models, offload_models=offload_models, device=device)
tokenizer_config = self.parse_path_or_model_id(tokenizer_path, ModelConfig(model_id="Qwen/Qwen3-0.6B", origin_file_pattern="./"))
tokenizer_t5xxl_config = self.parse_path_or_model_id(tokenizer_t5xxl_path, ModelConfig(model_id="stabilityai/stable-diffusion-3.5-large", origin_file_pattern="tokenizer_3/"))
self.pipe = AnimaImagePipeline.from_pretrained(torch_dtype=torch.bfloat16, device=device, model_configs=model_configs, tokenizer_config=tokenizer_config, tokenizer_t5xxl_config=tokenizer_t5xxl_config)
self.pipe = self.split_pipeline_units(task, self.pipe, trainable_models, lora_base_model)
# Training mode
self.switch_pipe_to_training_mode(
self.pipe, trainable_models,
lora_base_model, lora_target_modules, lora_rank, lora_checkpoint,
preset_lora_path, preset_lora_model,
task=task,
)
# Other configs
self.use_gradient_checkpointing = use_gradient_checkpointing
self.use_gradient_checkpointing_offload = use_gradient_checkpointing_offload
self.extra_inputs = extra_inputs.split(",") if extra_inputs is not None else []
self.fp8_models = fp8_models
self.task = task
self.task_to_loss = {
"sft:data_process": lambda pipe, *args: args,
"direct_distill:data_process": lambda pipe, *args: args,
"sft": lambda pipe, inputs_shared, inputs_posi, inputs_nega: FlowMatchSFTLoss(pipe, **inputs_shared, **inputs_posi),
"sft:train": lambda pipe, inputs_shared, inputs_posi, inputs_nega: FlowMatchSFTLoss(pipe, **inputs_shared, **inputs_posi),
"direct_distill": lambda pipe, inputs_shared, inputs_posi, inputs_nega: DirectDistillLoss(pipe, **inputs_shared, **inputs_posi),
"direct_distill:train": lambda pipe, inputs_shared, inputs_posi, inputs_nega: DirectDistillLoss(pipe, **inputs_shared, **inputs_posi),
}
def get_pipeline_inputs(self, data):
inputs_posi = {"prompt": data["prompt"]}
inputs_nega = {"negative_prompt": ""}
inputs_shared = {
# Assume you are using this pipeline for inference,
# please fill in the input parameters.
"input_image": data["image"],
"height": data["image"].size[1],
"width": data["image"].size[0],
# Please do not modify the following parameters
# unless you clearly know what this will cause.
"cfg_scale": 1,
"rand_device": self.pipe.device,
"use_gradient_checkpointing": self.use_gradient_checkpointing,
"use_gradient_checkpointing_offload": self.use_gradient_checkpointing_offload,
}
inputs_shared = self.parse_extra_inputs(data, self.extra_inputs, inputs_shared)
return inputs_shared, inputs_posi, inputs_nega
def forward(self, data, inputs=None):
if inputs is None: inputs = self.get_pipeline_inputs(data)
inputs = self.transfer_data_to_device(inputs, self.pipe.device, self.pipe.torch_dtype)
for unit in self.pipe.units:
inputs = self.pipe.unit_runner(unit, self.pipe, *inputs)
loss = self.task_to_loss[self.task](self.pipe, *inputs)
return loss
def anima_parser():
parser = argparse.ArgumentParser(description="Training script for Anima models.")
parser = add_general_config(parser)
parser = add_image_size_config(parser)
parser.add_argument("--tokenizer_path", type=str, default=None, help="Path to tokenizer.")
parser.add_argument("--tokenizer_t5xxl_path", type=str, default=None, help="Path to tokenizer_t5xxl.")
return parser
if __name__ == "__main__":
parser = anima_parser()
args = parser.parse_args()
accelerator = accelerate.Accelerator(
gradient_accumulation_steps=args.gradient_accumulation_steps,
kwargs_handlers=[accelerate.DistributedDataParallelKwargs(find_unused_parameters=args.find_unused_parameters)],
)
dataset = UnifiedDataset(
base_path=args.dataset_base_path,
metadata_path=args.dataset_metadata_path,
repeat=args.dataset_repeat,
data_file_keys=args.data_file_keys.split(","),
main_data_operator=UnifiedDataset.default_image_operator(
base_path=args.dataset_base_path,
max_pixels=args.max_pixels,
height=args.height,
width=args.width,
height_division_factor=16,
width_division_factor=16,
)
)
model = AnimaTrainingModule(
model_paths=args.model_paths,
model_id_with_origin_paths=args.model_id_with_origin_paths,
tokenizer_path=args.tokenizer_path,
tokenizer_t5xxl_path=args.tokenizer_t5xxl_path,
trainable_models=args.trainable_models,
lora_base_model=args.lora_base_model,
lora_target_modules=args.lora_target_modules,
lora_rank=args.lora_rank,
lora_checkpoint=args.lora_checkpoint,
preset_lora_path=args.preset_lora_path,
preset_lora_model=args.preset_lora_model,
use_gradient_checkpointing=args.use_gradient_checkpointing,
use_gradient_checkpointing_offload=args.use_gradient_checkpointing_offload,
extra_inputs=args.extra_inputs,
fp8_models=args.fp8_models,
offload_models=args.offload_models,
task=args.task,
device=accelerator.device,
)
model_logger = ModelLogger(
args.output_path,
remove_prefix_in_ckpt=args.remove_prefix_in_ckpt,
)
launcher_map = {
"sft:data_process": launch_data_process_task,
"direct_distill:data_process": launch_data_process_task,
"sft": launch_training_task,
"sft:train": launch_training_task,
"direct_distill": launch_training_task,
"direct_distill:train": launch_training_task,
}
launcher_map[args.task](accelerator, dataset, model, model_logger, args=args)

21
examples/anima/model_training/validate_full/anima-preview.py Normal file
View File

@@ -0,0 +1,21 @@
from diffsynth.pipelines.anima_image import AnimaImagePipeline, ModelConfig
from diffsynth.core import load_state_dict
import torch
pipe = AnimaImagePipeline.from_pretrained(
torch_dtype=torch.bfloat16,
device="cuda",
model_configs=[
ModelConfig(model_id="circlestone-labs/Anima", origin_file_pattern="split_files/diffusion_models/anima-preview.safetensors"),
ModelConfig(model_id="circlestone-labs/Anima", origin_file_pattern="split_files/text_encoders/qwen_3_06b_base.safetensors"),
ModelConfig(model_id="circlestone-labs/Anima", origin_file_pattern="split_files/vae/qwen_image_vae.safetensors"),
],
tokenizer_config=ModelConfig(model_id="Qwen/Qwen3-0.6B", origin_file_pattern="./"),
tokenizer_t5xxl_config=ModelConfig(model_id="stabilityai/stable-diffusion-3.5-large", origin_file_pattern="tokenizer_3/")
)
state_dict = load_state_dict("./models/train/anima-preview_full/epoch-1.safetensors", torch_dtype=torch.bfloat16)
pipe.dit.load_state_dict(state_dict)
prompt = "a dog"
image = pipe(prompt=prompt, seed=0)
image.save("image.jpg")

19
examples/anima/model_training/validate_lora/anima-preview.py Normal file
View File

@@ -0,0 +1,19 @@
from diffsynth.pipelines.anima_image import AnimaImagePipeline, ModelConfig
import torch
pipe = AnimaImagePipeline.from_pretrained(
torch_dtype=torch.bfloat16,
device="cuda",
model_configs=[
ModelConfig(model_id="circlestone-labs/Anima", origin_file_pattern="split_files/diffusion_models/anima-preview.safetensors"),
ModelConfig(model_id="circlestone-labs/Anima", origin_file_pattern="split_files/text_encoders/qwen_3_06b_base.safetensors"),
ModelConfig(model_id="circlestone-labs/Anima", origin_file_pattern="split_files/vae/qwen_image_vae.safetensors"),
],
tokenizer_config=ModelConfig(model_id="Qwen/Qwen3-0.6B", origin_file_pattern="./"),
tokenizer_t5xxl_config=ModelConfig(model_id="stabilityai/stable-diffusion-3.5-large", origin_file_pattern="tokenizer_3/")
)
pipe.load_lora(pipe.dit, "./models/train/anima-preview_lora/epoch-4.safetensors")
prompt = "a dog"
image = pipe(prompt=prompt, seed=0)
image.save("image.jpg")

7
examples/ltx2/model_inference/LTX-2-I2AV-DistilledPipeline.py
View File

@@ -19,7 +19,12 @@ pipe = LTX2AudioVideoPipeline.from_pretrained(
device="cuda",
model_configs=[
ModelConfig(model_id="google/gemma-3-12b-it-qat-q4_0-unquantized", origin_file_pattern="model-*.safetensors", **vram_config),
ModelConfig(model_id="Lightricks/LTX-2", origin_file_pattern="ltx-2-19b-distilled.safetensors", **vram_config),
ModelConfig(model_id="DiffSynth-Studio/LTX-2-Repackage", origin_file_pattern="transformer_distilled.safetensors", **vram_config),
ModelConfig(model_id="DiffSynth-Studio/LTX-2-Repackage", origin_file_pattern="text_encoder_post_modules.safetensors", **vram_config),
ModelConfig(model_id="DiffSynth-Studio/LTX-2-Repackage", origin_file_pattern="video_vae_decoder.safetensors", **vram_config),
ModelConfig(model_id="DiffSynth-Studio/LTX-2-Repackage", origin_file_pattern="audio_vae_decoder.safetensors", **vram_config),
ModelConfig(model_id="DiffSynth-Studio/LTX-2-Repackage", origin_file_pattern="audio_vocoder.safetensors", **vram_config),
ModelConfig(model_id="DiffSynth-Studio/LTX-2-Repackage", origin_file_pattern="video_vae_encoder.safetensors", **vram_config),
ModelConfig(model_id="Lightricks/LTX-2", origin_file_pattern="ltx-2-spatial-upscaler-x2-1.0.safetensors", **vram_config),
],
tokenizer_config=ModelConfig(model_id="google/gemma-3-12b-it-qat-q4_0-unquantized"),

7
examples/ltx2/model_inference/LTX-2-I2AV-OneStage.py
View File

@@ -19,7 +19,12 @@ pipe = LTX2AudioVideoPipeline.from_pretrained(
device="cuda",
model_configs=[
ModelConfig(model_id="google/gemma-3-12b-it-qat-q4_0-unquantized", origin_file_pattern="model-*.safetensors", **vram_config),
ModelConfig(model_id="Lightricks/LTX-2", origin_file_pattern="ltx-2-19b-dev.safetensors", **vram_config),
ModelConfig(model_id="DiffSynth-Studio/LTX-2-Repackage", origin_file_pattern="transformer.safetensors", **vram_config),
ModelConfig(model_id="DiffSynth-Studio/LTX-2-Repackage", origin_file_pattern="text_encoder_post_modules.safetensors", **vram_config),
ModelConfig(model_id="DiffSynth-Studio/LTX-2-Repackage", origin_file_pattern="video_vae_decoder.safetensors", **vram_config),
ModelConfig(model_id="DiffSynth-Studio/LTX-2-Repackage", origin_file_pattern="audio_vae_decoder.safetensors", **vram_config),
ModelConfig(model_id="DiffSynth-Studio/LTX-2-Repackage", origin_file_pattern="audio_vocoder.safetensors", **vram_config),
ModelConfig(model_id="DiffSynth-Studio/LTX-2-Repackage", origin_file_pattern="video_vae_encoder.safetensors", **vram_config),
],
tokenizer_config=ModelConfig(model_id="google/gemma-3-12b-it-qat-q4_0-unquantized"),
)

8
examples/ltx2/model_inference/LTX-2-I2AV-TwoStage.py
View File

@@ -19,7 +19,12 @@ pipe = LTX2AudioVideoPipeline.from_pretrained(
device="cuda",
model_configs=[
ModelConfig(model_id="google/gemma-3-12b-it-qat-q4_0-unquantized", origin_file_pattern="model-*.safetensors", **vram_config),
ModelConfig(model_id="Lightricks/LTX-2", origin_file_pattern="ltx-2-19b-dev.safetensors", **vram_config),
ModelConfig(model_id="DiffSynth-Studio/LTX-2-Repackage", origin_file_pattern="transformer.safetensors", **vram_config),
ModelConfig(model_id="DiffSynth-Studio/LTX-2-Repackage", origin_file_pattern="text_encoder_post_modules.safetensors", **vram_config),
ModelConfig(model_id="DiffSynth-Studio/LTX-2-Repackage", origin_file_pattern="video_vae_decoder.safetensors", **vram_config),
ModelConfig(model_id="DiffSynth-Studio/LTX-2-Repackage", origin_file_pattern="audio_vae_decoder.safetensors", **vram_config),
ModelConfig(model_id="DiffSynth-Studio/LTX-2-Repackage", origin_file_pattern="audio_vocoder.safetensors", **vram_config),
ModelConfig(model_id="DiffSynth-Studio/LTX-2-Repackage", origin_file_pattern="video_vae_encoder.safetensors", **vram_config),
ModelConfig(model_id="Lightricks/LTX-2", origin_file_pattern="ltx-2-spatial-upscaler-x2-1.0.safetensors", **vram_config),
],
tokenizer_config=ModelConfig(model_id="google/gemma-3-12b-it-qat-q4_0-unquantized"),
@@ -41,7 +46,6 @@ negative_prompt = (
"inconsistent tone, cinematic oversaturation, stylized filters, or AI artifacts."
)
height, width, num_frames = 512 * 2, 768 * 2, 121
height, width, num_frames = 512 * 2, 768 * 2, 121
dataset_snapshot_download(
dataset_id="DiffSynth-Studio/examples_in_diffsynth",
local_dir="./",

7
examples/ltx2/model_inference/LTX-2-T2AV-Camera-Control-Dolly-In.py
View File

@@ -17,7 +17,12 @@ pipe = LTX2AudioVideoPipeline.from_pretrained(
device="cuda",
model_configs=[
ModelConfig(model_id="google/gemma-3-12b-it-qat-q4_0-unquantized", origin_file_pattern="model-*.safetensors", **vram_config),
ModelConfig(model_id="Lightricks/LTX-2", origin_file_pattern="ltx-2-19b-dev.safetensors", **vram_config),
ModelConfig(model_id="DiffSynth-Studio/LTX-2-Repackage", origin_file_pattern="transformer.safetensors", **vram_config),
ModelConfig(model_id="DiffSynth-Studio/LTX-2-Repackage", origin_file_pattern="text_encoder_post_modules.safetensors", **vram_config),
ModelConfig(model_id="DiffSynth-Studio/LTX-2-Repackage", origin_file_pattern="video_vae_decoder.safetensors", **vram_config),
ModelConfig(model_id="DiffSynth-Studio/LTX-2-Repackage", origin_file_pattern="audio_vae_decoder.safetensors", **vram_config),
ModelConfig(model_id="DiffSynth-Studio/LTX-2-Repackage", origin_file_pattern="audio_vocoder.safetensors", **vram_config),
ModelConfig(model_id="DiffSynth-Studio/LTX-2-Repackage", origin_file_pattern="video_vae_encoder.safetensors", **vram_config),
ModelConfig(model_id="Lightricks/LTX-2", origin_file_pattern="ltx-2-spatial-upscaler-x2-1.0.safetensors", **vram_config),
],
tokenizer_config=ModelConfig(model_id="google/gemma-3-12b-it-qat-q4_0-unquantized"),

7
examples/ltx2/model_inference/LTX-2-T2AV-Camera-Control-Dolly-Left.py
View File

@@ -17,7 +17,12 @@ pipe = LTX2AudioVideoPipeline.from_pretrained(
device="cuda",
model_configs=[
ModelConfig(model_id="google/gemma-3-12b-it-qat-q4_0-unquantized", origin_file_pattern="model-*.safetensors", **vram_config),
ModelConfig(model_id="Lightricks/LTX-2", origin_file_pattern="ltx-2-19b-dev.safetensors", **vram_config),
ModelConfig(model_id="DiffSynth-Studio/LTX-2-Repackage", origin_file_pattern="transformer.safetensors", **vram_config),
ModelConfig(model_id="DiffSynth-Studio/LTX-2-Repackage", origin_file_pattern="text_encoder_post_modules.safetensors", **vram_config),
ModelConfig(model_id="DiffSynth-Studio/LTX-2-Repackage", origin_file_pattern="video_vae_decoder.safetensors", **vram_config),
ModelConfig(model_id="DiffSynth-Studio/LTX-2-Repackage", origin_file_pattern="audio_vae_decoder.safetensors", **vram_config),
ModelConfig(model_id="DiffSynth-Studio/LTX-2-Repackage", origin_file_pattern="audio_vocoder.safetensors", **vram_config),
ModelConfig(model_id="DiffSynth-Studio/LTX-2-Repackage", origin_file_pattern="video_vae_encoder.safetensors", **vram_config),
ModelConfig(model_id="Lightricks/LTX-2", origin_file_pattern="ltx-2-spatial-upscaler-x2-1.0.safetensors", **vram_config),
],
tokenizer_config=ModelConfig(model_id="google/gemma-3-12b-it-qat-q4_0-unquantized"),

7
examples/ltx2/model_inference/LTX-2-T2AV-Camera-Control-Dolly-Out.py
View File

@@ -17,7 +17,12 @@ pipe = LTX2AudioVideoPipeline.from_pretrained(
device="cuda",
model_configs=[
ModelConfig(model_id="google/gemma-3-12b-it-qat-q4_0-unquantized", origin_file_pattern="model-*.safetensors", **vram_config),
ModelConfig(model_id="Lightricks/LTX-2", origin_file_pattern="ltx-2-19b-dev.safetensors", **vram_config),
ModelConfig(model_id="DiffSynth-Studio/LTX-2-Repackage", origin_file_pattern="transformer.safetensors", **vram_config),
ModelConfig(model_id="DiffSynth-Studio/LTX-2-Repackage", origin_file_pattern="text_encoder_post_modules.safetensors", **vram_config),
ModelConfig(model_id="DiffSynth-Studio/LTX-2-Repackage", origin_file_pattern="video_vae_decoder.safetensors", **vram_config),
ModelConfig(model_id="DiffSynth-Studio/LTX-2-Repackage", origin_file_pattern="audio_vae_decoder.safetensors", **vram_config),
ModelConfig(model_id="DiffSynth-Studio/LTX-2-Repackage", origin_file_pattern="audio_vocoder.safetensors", **vram_config),
ModelConfig(model_id="DiffSynth-Studio/LTX-2-Repackage", origin_file_pattern="video_vae_encoder.safetensors", **vram_config),
ModelConfig(model_id="Lightricks/LTX-2", origin_file_pattern="ltx-2-spatial-upscaler-x2-1.0.safetensors", **vram_config),
],
tokenizer_config=ModelConfig(model_id="google/gemma-3-12b-it-qat-q4_0-unquantized"),

7
examples/ltx2/model_inference/LTX-2-T2AV-Camera-Control-Dolly-Right.py
View File

@@ -17,7 +17,12 @@ pipe = LTX2AudioVideoPipeline.from_pretrained(
device="cuda",
model_configs=[
ModelConfig(model_id="google/gemma-3-12b-it-qat-q4_0-unquantized", origin_file_pattern="model-*.safetensors", **vram_config),
ModelConfig(model_id="Lightricks/LTX-2", origin_file_pattern="ltx-2-19b-dev.safetensors", **vram_config),
ModelConfig(model_id="DiffSynth-Studio/LTX-2-Repackage", origin_file_pattern="transformer.safetensors", **vram_config),
ModelConfig(model_id="DiffSynth-Studio/LTX-2-Repackage", origin_file_pattern="text_encoder_post_modules.safetensors", **vram_config),
ModelConfig(model_id="DiffSynth-Studio/LTX-2-Repackage", origin_file_pattern="video_vae_decoder.safetensors", **vram_config),
ModelConfig(model_id="DiffSynth-Studio/LTX-2-Repackage", origin_file_pattern="audio_vae_decoder.safetensors", **vram_config),
ModelConfig(model_id="DiffSynth-Studio/LTX-2-Repackage", origin_file_pattern="audio_vocoder.safetensors", **vram_config),
ModelConfig(model_id="DiffSynth-Studio/LTX-2-Repackage", origin_file_pattern="video_vae_encoder.safetensors", **vram_config),
ModelConfig(model_id="Lightricks/LTX-2", origin_file_pattern="ltx-2-spatial-upscaler-x2-1.0.safetensors", **vram_config),
],
tokenizer_config=ModelConfig(model_id="google/gemma-3-12b-it-qat-q4_0-unquantized"),

7
examples/ltx2/model_inference/LTX-2-T2AV-Camera-Control-Jib-Down.py
View File

@@ -17,7 +17,12 @@ pipe = LTX2AudioVideoPipeline.from_pretrained(
device="cuda",
model_configs=[
ModelConfig(model_id="google/gemma-3-12b-it-qat-q4_0-unquantized", origin_file_pattern="model-*.safetensors", **vram_config),
ModelConfig(model_id="Lightricks/LTX-2", origin_file_pattern="ltx-2-19b-dev.safetensors", **vram_config),
ModelConfig(model_id="DiffSynth-Studio/LTX-2-Repackage", origin_file_pattern="transformer.safetensors", **vram_config),
ModelConfig(model_id="DiffSynth-Studio/LTX-2-Repackage", origin_file_pattern="text_encoder_post_modules.safetensors", **vram_config),
ModelConfig(model_id="DiffSynth-Studio/LTX-2-Repackage", origin_file_pattern="video_vae_decoder.safetensors", **vram_config),
ModelConfig(model_id="DiffSynth-Studio/LTX-2-Repackage", origin_file_pattern="audio_vae_decoder.safetensors", **vram_config),
ModelConfig(model_id="DiffSynth-Studio/LTX-2-Repackage", origin_file_pattern="audio_vocoder.safetensors", **vram_config),
ModelConfig(model_id="DiffSynth-Studio/LTX-2-Repackage", origin_file_pattern="video_vae_encoder.safetensors", **vram_config),
ModelConfig(model_id="Lightricks/LTX-2", origin_file_pattern="ltx-2-spatial-upscaler-x2-1.0.safetensors", **vram_config),
],
tokenizer_config=ModelConfig(model_id="google/gemma-3-12b-it-qat-q4_0-unquantized"),

7
examples/ltx2/model_inference/LTX-2-T2AV-Camera-Control-Jib-Up.py
View File

@@ -17,7 +17,12 @@ pipe = LTX2AudioVideoPipeline.from_pretrained(
device="cuda",
model_configs=[
ModelConfig(model_id="google/gemma-3-12b-it-qat-q4_0-unquantized", origin_file_pattern="model-*.safetensors", **vram_config),
ModelConfig(model_id="Lightricks/LTX-2", origin_file_pattern="ltx-2-19b-dev.safetensors", **vram_config),
ModelConfig(model_id="DiffSynth-Studio/LTX-2-Repackage", origin_file_pattern="transformer.safetensors", **vram_config),
ModelConfig(model_id="DiffSynth-Studio/LTX-2-Repackage", origin_file_pattern="text_encoder_post_modules.safetensors", **vram_config),
ModelConfig(model_id="DiffSynth-Studio/LTX-2-Repackage", origin_file_pattern="video_vae_decoder.safetensors", **vram_config),
ModelConfig(model_id="DiffSynth-Studio/LTX-2-Repackage", origin_file_pattern="audio_vae_decoder.safetensors", **vram_config),
ModelConfig(model_id="DiffSynth-Studio/LTX-2-Repackage", origin_file_pattern="audio_vocoder.safetensors", **vram_config),
ModelConfig(model_id="DiffSynth-Studio/LTX-2-Repackage", origin_file_pattern="video_vae_encoder.safetensors", **vram_config),
ModelConfig(model_id="Lightricks/LTX-2", origin_file_pattern="ltx-2-spatial-upscaler-x2-1.0.safetensors", **vram_config),
],
tokenizer_config=ModelConfig(model_id="google/gemma-3-12b-it-qat-q4_0-unquantized"),

7
examples/ltx2/model_inference/LTX-2-T2AV-Camera-Control-Static.py
View File

@@ -17,7 +17,12 @@ pipe = LTX2AudioVideoPipeline.from_pretrained(
device="cuda",
model_configs=[
ModelConfig(model_id="google/gemma-3-12b-it-qat-q4_0-unquantized", origin_file_pattern="model-*.safetensors", **vram_config),
ModelConfig(model_id="Lightricks/LTX-2", origin_file_pattern="ltx-2-19b-dev.safetensors", **vram_config),
ModelConfig(model_id="DiffSynth-Studio/LTX-2-Repackage", origin_file_pattern="transformer.safetensors", **vram_config),
ModelConfig(model_id="DiffSynth-Studio/LTX-2-Repackage", origin_file_pattern="text_encoder_post_modules.safetensors", **vram_config),
ModelConfig(model_id="DiffSynth-Studio/LTX-2-Repackage", origin_file_pattern="video_vae_decoder.safetensors", **vram_config),
ModelConfig(model_id="DiffSynth-Studio/LTX-2-Repackage", origin_file_pattern="audio_vae_decoder.safetensors", **vram_config),
ModelConfig(model_id="DiffSynth-Studio/LTX-2-Repackage", origin_file_pattern="audio_vocoder.safetensors", **vram_config),
ModelConfig(model_id="DiffSynth-Studio/LTX-2-Repackage", origin_file_pattern="video_vae_encoder.safetensors", **vram_config),
ModelConfig(model_id="Lightricks/LTX-2", origin_file_pattern="ltx-2-spatial-upscaler-x2-1.0.safetensors", **vram_config),
],
tokenizer_config=ModelConfig(model_id="google/gemma-3-12b-it-qat-q4_0-unquantized"),

7
examples/ltx2/model_inference/LTX-2-T2AV-DistilledPipeline.py
View File

@@ -17,7 +17,12 @@ pipe = LTX2AudioVideoPipeline.from_pretrained(
device="cuda",
model_configs=[
ModelConfig(model_id="google/gemma-3-12b-it-qat-q4_0-unquantized", origin_file_pattern="model-*.safetensors", **vram_config),
ModelConfig(model_id="Lightricks/LTX-2", origin_file_pattern="ltx-2-19b-distilled.safetensors", **vram_config),
ModelConfig(model_id="DiffSynth-Studio/LTX-2-Repackage", origin_file_pattern="transformer_distilled.safetensors", **vram_config),
ModelConfig(model_id="DiffSynth-Studio/LTX-2-Repackage", origin_file_pattern="text_encoder_post_modules.safetensors", **vram_config),
ModelConfig(model_id="DiffSynth-Studio/LTX-2-Repackage", origin_file_pattern="video_vae_decoder.safetensors", **vram_config),
ModelConfig(model_id="DiffSynth-Studio/LTX-2-Repackage", origin_file_pattern="audio_vae_decoder.safetensors", **vram_config),
ModelConfig(model_id="DiffSynth-Studio/LTX-2-Repackage", origin_file_pattern="audio_vocoder.safetensors", **vram_config),
ModelConfig(model_id="DiffSynth-Studio/LTX-2-Repackage", origin_file_pattern="video_vae_encoder.safetensors", **vram_config),
ModelConfig(model_id="Lightricks/LTX-2", origin_file_pattern="ltx-2-spatial-upscaler-x2-1.0.safetensors", **vram_config),
],
tokenizer_config=ModelConfig(model_id="google/gemma-3-12b-it-qat-q4_0-unquantized"),

77
examples/ltx2/model_inference/LTX-2-T2AV-IC-LoRA-Detailer.py Normal file
View File

@@ -0,0 +1,77 @@
import torch
from diffsynth.pipelines.ltx2_audio_video import LTX2AudioVideoPipeline, ModelConfig
from diffsynth.utils.data.media_io_ltx2 import write_video_audio_ltx2
from diffsynth.utils.data import VideoData
from modelscope import dataset_snapshot_download
vram_config = {
"offload_dtype": torch.bfloat16,
"offload_device": "cpu",
"onload_dtype": torch.bfloat16,
"onload_device": "cuda",
"preparing_dtype": torch.bfloat16,
"preparing_device": "cuda",
"computation_dtype": torch.bfloat16,
"computation_device": "cuda",
}
pipe = LTX2AudioVideoPipeline.from_pretrained(
torch_dtype=torch.bfloat16,
device="cuda",
model_configs=[
ModelConfig(model_id="google/gemma-3-12b-it-qat-q4_0-unquantized", origin_file_pattern="model-*.safetensors", **vram_config),
ModelConfig(model_id="DiffSynth-Studio/LTX-2-Repackage", origin_file_pattern="transformer.safetensors", **vram_config),
ModelConfig(model_id="DiffSynth-Studio/LTX-2-Repackage", origin_file_pattern="text_encoder_post_modules.safetensors", **vram_config),
ModelConfig(model_id="DiffSynth-Studio/LTX-2-Repackage", origin_file_pattern="video_vae_decoder.safetensors", **vram_config),
ModelConfig(model_id="DiffSynth-Studio/LTX-2-Repackage", origin_file_pattern="audio_vae_decoder.safetensors", **vram_config),
ModelConfig(model_id="DiffSynth-Studio/LTX-2-Repackage", origin_file_pattern="audio_vocoder.safetensors", **vram_config),
ModelConfig(model_id="DiffSynth-Studio/LTX-2-Repackage", origin_file_pattern="video_vae_encoder.safetensors", **vram_config),
ModelConfig(model_id="Lightricks/LTX-2", origin_file_pattern="ltx-2-spatial-upscaler-x2-1.0.safetensors", **vram_config),
],
tokenizer_config=ModelConfig(model_id="google/gemma-3-12b-it-qat-q4_0-unquantized"),
stage2_lora_config=ModelConfig(model_id="Lightricks/LTX-2", origin_file_pattern="ltx-2-19b-distilled-lora-384.safetensors"),
)
pipe.load_lora(pipe.dit, ModelConfig(model_id="Lightricks/LTX-2-19b-IC-LoRA-Detailer", origin_file_pattern="ltx-2-19b-ic-lora-detailer.safetensors"))
dataset_snapshot_download("DiffSynth-Studio/example_video_dataset", allow_file_pattern="ltx2/*", local_dir="data/example_video_dataset")
prompt = "[VISUAL]:Two cute orange cats, wearing boxing gloves, stand on a boxing ring and fight each other. [SOUNDS]:the sound of two cats boxing"
negative_prompt = (
"blurry, out of focus, overexposed, underexposed, low contrast, washed out colors, excessive noise, "
"grainy texture, poor lighting, flickering, motion blur, distorted proportions, unnatural skin tones, "
"deformed facial features, asymmetrical face, missing facial features, extra limbs, disfigured hands, "
"wrong hand count, artifacts around text, inconsistent perspective, camera shake, incorrect depth of "
"field, background too sharp, background clutter, distracting reflections, harsh shadows, inconsistent "
"lighting direction, color banding, cartoonish rendering, 3D CGI look, unrealistic materials, uncanny "
"valley effect, incorrect ethnicity, wrong gender, exaggerated expressions, wrong gaze direction, "
"mismatched lip sync, silent or muted audio, distorted voice, robotic voice, echo, background noise, "
"off-sync audio, incorrect dialogue, added dialogue, repetitive speech, jittery movement, awkward "
"pauses, incorrect timing, unnatural transitions, inconsistent framing, tilted camera, flat lighting, "
"inconsistent tone, cinematic oversaturation, stylized filters, or AI artifacts."
)
height, width, num_frames = 512 * 2, 768 * 2, 121
ref_scale_factor = 1
frame_rate = 24
# the frame rate of the video should better be the same with the reference video
# the spatial resolution of the first frame should be the resolution of stage 1 video generation divided by ref_scale_factor
input_video = VideoData("data/example_video_dataset/ltx2/video1.mp4", height=height // ref_scale_factor // 2, width=width // ref_scale_factor // 2)
input_video = input_video.raw_data()
video, audio = pipe(
prompt=prompt,
negative_prompt=negative_prompt,
seed=43,
height=height,
width=width,
num_frames=num_frames,
frame_rate=frame_rate,
in_context_videos=[input_video],
in_context_downsample_factor=ref_scale_factor,
tiled=True,
use_two_stage_pipeline=True,
clear_lora_before_state_two=True,
)
write_video_audio_ltx2(
video=video,
audio=audio,
output_path='ltx2_twostage_iclora.mp4',
fps=frame_rate,
audio_sample_rate=24000,
)

77
examples/ltx2/model_inference/LTX-2-T2AV-IC-LoRA-Union-Control.py Normal file
View File

@@ -0,0 +1,77 @@
import torch
from diffsynth.pipelines.ltx2_audio_video import LTX2AudioVideoPipeline, ModelConfig
from diffsynth.utils.data.media_io_ltx2 import write_video_audio_ltx2
from diffsynth.utils.data import VideoData
from modelscope import dataset_snapshot_download
vram_config = {
"offload_dtype": torch.bfloat16,
"offload_device": "cpu",
"onload_dtype": torch.bfloat16,
"onload_device": "cuda",
"preparing_dtype": torch.bfloat16,
"preparing_device": "cuda",
"computation_dtype": torch.bfloat16,
"computation_device": "cuda",
}
pipe = LTX2AudioVideoPipeline.from_pretrained(
torch_dtype=torch.bfloat16,
device="cuda",
model_configs=[
ModelConfig(model_id="google/gemma-3-12b-it-qat-q4_0-unquantized", origin_file_pattern="model-*.safetensors", **vram_config),
ModelConfig(model_id="DiffSynth-Studio/LTX-2-Repackage", origin_file_pattern="transformer.safetensors", **vram_config),
ModelConfig(model_id="DiffSynth-Studio/LTX-2-Repackage", origin_file_pattern="text_encoder_post_modules.safetensors", **vram_config),
ModelConfig(model_id="DiffSynth-Studio/LTX-2-Repackage", origin_file_pattern="video_vae_decoder.safetensors", **vram_config),
ModelConfig(model_id="DiffSynth-Studio/LTX-2-Repackage", origin_file_pattern="audio_vae_decoder.safetensors", **vram_config),
ModelConfig(model_id="DiffSynth-Studio/LTX-2-Repackage", origin_file_pattern="audio_vocoder.safetensors", **vram_config),
ModelConfig(model_id="DiffSynth-Studio/LTX-2-Repackage", origin_file_pattern="video_vae_encoder.safetensors", **vram_config),
ModelConfig(model_id="Lightricks/LTX-2", origin_file_pattern="ltx-2-spatial-upscaler-x2-1.0.safetensors", **vram_config),
],
tokenizer_config=ModelConfig(model_id="google/gemma-3-12b-it-qat-q4_0-unquantized"),
stage2_lora_config=ModelConfig(model_id="Lightricks/LTX-2", origin_file_pattern="ltx-2-19b-distilled-lora-384.safetensors"),
)
pipe.load_lora(pipe.dit, ModelConfig(model_id="Lightricks/LTX-2-19b-IC-LoRA-Union-Control", origin_file_pattern="ltx-2-19b-ic-lora-union-control-ref0.5.safetensors"))
dataset_snapshot_download("DiffSynth-Studio/example_video_dataset", allow_file_pattern="ltx2/*", local_dir="data/example_video_dataset")
prompt = "[VISUAL]:Two cute orange cats, wearing boxing gloves, stand on a boxing ring and fight each other. [SOUNDS]:the sound of two cats boxing"
negative_prompt = (
"blurry, out of focus, overexposed, underexposed, low contrast, washed out colors, excessive noise, "
"grainy texture, poor lighting, flickering, motion blur, distorted proportions, unnatural skin tones, "
"deformed facial features, asymmetrical face, missing facial features, extra limbs, disfigured hands, "
"wrong hand count, artifacts around text, inconsistent perspective, camera shake, incorrect depth of "
"field, background too sharp, background clutter, distracting reflections, harsh shadows, inconsistent "
"lighting direction, color banding, cartoonish rendering, 3D CGI look, unrealistic materials, uncanny "
"valley effect, incorrect ethnicity, wrong gender, exaggerated expressions, wrong gaze direction, "
"mismatched lip sync, silent or muted audio, distorted voice, robotic voice, echo, background noise, "
"off-sync audio, incorrect dialogue, added dialogue, repetitive speech, jittery movement, awkward "
"pauses, incorrect timing, unnatural transitions, inconsistent framing, tilted camera, flat lighting, "
"inconsistent tone, cinematic oversaturation, stylized filters, or AI artifacts."
)
height, width, num_frames = 512 * 2, 768 * 2, 121
ref_scale_factor = 2
frame_rate = 24
# the frame rate of the video should better be the same with the reference video
# the spatial resolution of the first frame should be the resolution of stage 1 video generation divided by ref_scale_factor
input_video = VideoData("data/example_video_dataset/ltx2/depth_video.mp4", height=height // ref_scale_factor // 2, width=width // ref_scale_factor // 2)
input_video = input_video.raw_data()
video, audio = pipe(
prompt=prompt,
negative_prompt=negative_prompt,
seed=43,
height=height,
width=width,
num_frames=num_frames,
frame_rate=frame_rate,
in_context_videos=[input_video],
in_context_downsample_factor=ref_scale_factor,
tiled=True,
use_two_stage_pipeline=True,
clear_lora_before_state_two=True,
)
write_video_audio_ltx2(
video=video,
audio=audio,
output_path='ltx2_twostage_iclora.mp4',
fps=frame_rate,
audio_sample_rate=24000,
)

25
examples/ltx2/model_inference/LTX-2-T2AV-OneStage.py
View File

@@ -12,15 +12,38 @@ vram_config = {
"computation_dtype": torch.bfloat16,
"computation_device": "cuda",
}
"""
Offical model repo: https://www.modelscope.cn/models/Lightricks/LTX-2
Repackaged model repo: https://www.modelscope.cn/models/DiffSynth-Studio/LTX-2-Repackage
For base models of LTX-2, offical checkpoint (with model config ModelConfig(model_id="Lightricks/LTX-2", origin_file_pattern="ltx-2-19b-dev.safetensors"))
and repackaged checkpoints (with model config ModelConfig(model_id="DiffSynth-Studio/LTX-2-Repackage", origin_file_pattern="*.safetensors")) are both supported.
We have repackeged the official checkpoints in DiffSynth-Studio/LTX-2-Repackage repo to support separate loading of different submodules,
and avoid redundant memory usage when users only want to use part of the model.
"""
# use the repackaged modelconfig from "DiffSynth-Studio/LTX-2-Repackage" to avoid redundant model loading
pipe = LTX2AudioVideoPipeline.from_pretrained(
torch_dtype=torch.bfloat16,
device="cuda",
model_configs=[
ModelConfig(model_id="google/gemma-3-12b-it-qat-q4_0-unquantized", origin_file_pattern="model-*.safetensors", **vram_config),
ModelConfig(model_id="Lightricks/LTX-2", origin_file_pattern="ltx-2-19b-dev.safetensors", **vram_config),
ModelConfig(model_id="DiffSynth-Studio/LTX-2-Repackage", origin_file_pattern="transformer.safetensors", **vram_config),
ModelConfig(model_id="DiffSynth-Studio/LTX-2-Repackage", origin_file_pattern="text_encoder_post_modules.safetensors", **vram_config),
ModelConfig(model_id="DiffSynth-Studio/LTX-2-Repackage", origin_file_pattern="video_vae_decoder.safetensors", **vram_config),
ModelConfig(model_id="DiffSynth-Studio/LTX-2-Repackage", origin_file_pattern="audio_vae_decoder.safetensors", **vram_config),
ModelConfig(model_id="DiffSynth-Studio/LTX-2-Repackage", origin_file_pattern="audio_vocoder.safetensors", **vram_config),
],
tokenizer_config=ModelConfig(model_id="google/gemma-3-12b-it-qat-q4_0-unquantized"),
)
# use the following modelconfig if you want to initialize model from offical checkpoints from "Lightricks/LTX-2"
# pipe = LTX2AudioVideoPipeline.from_pretrained(
# torch_dtype=torch.bfloat16,
# device="cuda",
# model_configs=[
# ModelConfig(model_id="google/gemma-3-12b-it-qat-q4_0-unquantized", origin_file_pattern="model-*.safetensors", **vram_config),
# ModelConfig(model_id="Lightricks/LTX-2", origin_file_pattern="ltx-2-19b-dev.safetensors", **vram_config),
# ],
# tokenizer_config=ModelConfig(model_id="google/gemma-3-12b-it-qat-q4_0-unquantized"),
# )
prompt = "A girl is very happy, she is speaking: “I enjoy working with Diffsynth-Studio, it's a perfect framework.”"
negative_prompt = "blurry, out of focus, overexposed, underexposed, low contrast, washed out colors, excessive noise, grainy texture, poor lighting, flickering, motion blur, distorted proportions, unnatural skin tones, deformed facial features, asymmetrical face, missing facial features, extra limbs, disfigured hands, wrong hand count, artifacts around text, inconsistent perspective, camera shake, incorrect depth of field, background too sharp, background clutter, distracting reflections, harsh shadows, inconsistent lighting direction, color banding, cartoonish rendering, 3D CGI look, unrealistic materials, uncanny valley effect, incorrect ethnicity, wrong gender, exaggerated expressions, wrong gaze direction, mismatched lip sync, silent or muted audio, distorted voice, robotic voice, echo, background noise, off-sync audio, incorrect dialogue, added dialogue, repetitive speech, jittery movement, awkward pauses, incorrect timing, unnatural transitions, inconsistent framing, tilted camera, flat lighting, inconsistent tone, cinematic oversaturation, stylized filters, or AI artifacts."
height, width, num_frames = 512, 768, 121

29
examples/ltx2/model_inference/LTX-2-T2AV-TwoStage.py
View File

@@ -12,18 +12,43 @@ vram_config = {
"computation_dtype": torch.bfloat16,
"computation_device": "cuda",
}
"""
Offical model repo: https://www.modelscope.cn/models/Lightricks/LTX-2
Repackaged model repo: https://www.modelscope.cn/models/DiffSynth-Studio/LTX-2-Repackage
For base models of LTX-2, offical checkpoint (with model config ModelConfig(model_id="Lightricks/LTX-2", origin_file_pattern="ltx-2-19b-dev.safetensors"))
and repackaged checkpoints (with model config ModelConfig(model_id="DiffSynth-Studio/LTX-2-Repackage", origin_file_pattern="*.safetensors")) are both supported.
We have repackeged the official checkpoints in DiffSynth-Studio/LTX-2-Repackage repo to support separate loading of different submodules,
and avoid redundant memory usage when users only want to use part of the model.
"""
# use the repackaged modelconfig from "DiffSynth-Studio/LTX-2-Repackage" to avoid redundant model loading
pipe = LTX2AudioVideoPipeline.from_pretrained(
torch_dtype=torch.bfloat16,
device="cuda",
model_configs=[
ModelConfig(model_id="google/gemma-3-12b-it-qat-q4_0-unquantized", origin_file_pattern="model-*.safetensors", **vram_config),
ModelConfig(model_id="Lightricks/LTX-2", origin_file_pattern="ltx-2-19b-dev.safetensors", **vram_config),
ModelConfig(model_id="DiffSynth-Studio/LTX-2-Repackage", origin_file_pattern="transformer.safetensors", **vram_config),
ModelConfig(model_id="DiffSynth-Studio/LTX-2-Repackage", origin_file_pattern="text_encoder_post_modules.safetensors", **vram_config),
ModelConfig(model_id="DiffSynth-Studio/LTX-2-Repackage", origin_file_pattern="video_vae_decoder.safetensors", **vram_config),
ModelConfig(model_id="DiffSynth-Studio/LTX-2-Repackage", origin_file_pattern="audio_vae_decoder.safetensors", **vram_config),
ModelConfig(model_id="DiffSynth-Studio/LTX-2-Repackage", origin_file_pattern="audio_vocoder.safetensors", **vram_config),
ModelConfig(model_id="DiffSynth-Studio/LTX-2-Repackage", origin_file_pattern="video_vae_encoder.safetensors", **vram_config),
ModelConfig(model_id="Lightricks/LTX-2", origin_file_pattern="ltx-2-spatial-upscaler-x2-1.0.safetensors", **vram_config),
],
tokenizer_config=ModelConfig(model_id="google/gemma-3-12b-it-qat-q4_0-unquantized"),
stage2_lora_config=ModelConfig(model_id="Lightricks/LTX-2", origin_file_pattern="ltx-2-19b-distilled-lora-384.safetensors"),
)
# use the following modelconfig if you want to initialize model from offical checkpoints from "Lightricks/LTX-2"
# pipe = LTX2AudioVideoPipeline.from_pretrained(
# torch_dtype=torch.bfloat16,
# device="cuda",
# model_configs=[
# ModelConfig(model_id="google/gemma-3-12b-it-qat-q4_0-unquantized", origin_file_pattern="model-*.safetensors", **vram_config),
# ModelConfig(model_id="Lightricks/LTX-2", origin_file_pattern="ltx-2-19b-dev.safetensors", **vram_config),
# ModelConfig(model_id="Lightricks/LTX-2", origin_file_pattern="ltx-2-spatial-upscaler-x2-1.0.safetensors", **vram_config),
# ],
# tokenizer_config=ModelConfig(model_id="google/gemma-3-12b-it-qat-q4_0-unquantized"),
# stage2_lora_config=ModelConfig(model_id="Lightricks/LTX-2", origin_file_pattern="ltx-2-19b-distilled-lora-384.safetensors"),
# )
prompt = "A girl is very happy, she is speaking: “I enjoy working with Diffsynth-Studio, it's a perfect framework.”"
negative_prompt = (
"blurry, out of focus, overexposed, underexposed, low contrast, washed out colors, excessive noise, "

7
examples/ltx2/model_inference_low_vram/LTX-2-I2AV-DistilledPipeline.py
View File

@@ -19,7 +19,12 @@ pipe = LTX2AudioVideoPipeline.from_pretrained(
device="cuda",
model_configs=[
ModelConfig(model_id="google/gemma-3-12b-it-qat-q4_0-unquantized", origin_file_pattern="model-*.safetensors", **vram_config),
ModelConfig(model_id="Lightricks/LTX-2", origin_file_pattern="ltx-2-19b-distilled.safetensors", **vram_config),
ModelConfig(model_id="DiffSynth-Studio/LTX-2-Repackage", origin_file_pattern="transformer_distilled.safetensors", **vram_config),
ModelConfig(model_id="DiffSynth-Studio/LTX-2-Repackage", origin_file_pattern="text_encoder_post_modules.safetensors", **vram_config),
ModelConfig(model_id="DiffSynth-Studio/LTX-2-Repackage", origin_file_pattern="video_vae_decoder.safetensors", **vram_config),
ModelConfig(model_id="DiffSynth-Studio/LTX-2-Repackage", origin_file_pattern="audio_vae_decoder.safetensors", **vram_config),
ModelConfig(model_id="DiffSynth-Studio/LTX-2-Repackage", origin_file_pattern="audio_vocoder.safetensors", **vram_config),
ModelConfig(model_id="DiffSynth-Studio/LTX-2-Repackage", origin_file_pattern="video_vae_encoder.safetensors", **vram_config),
ModelConfig(model_id="Lightricks/LTX-2", origin_file_pattern="ltx-2-spatial-upscaler-x2-1.0.safetensors", **vram_config),
],
tokenizer_config=ModelConfig(model_id="google/gemma-3-12b-it-qat-q4_0-unquantized"),

7
examples/ltx2/model_inference_low_vram/LTX-2-I2AV-OneStage.py
View File

@@ -19,7 +19,12 @@ pipe = LTX2AudioVideoPipeline.from_pretrained(
device="cuda",
model_configs=[
ModelConfig(model_id="google/gemma-3-12b-it-qat-q4_0-unquantized", origin_file_pattern="model-*.safetensors", **vram_config),
ModelConfig(model_id="Lightricks/LTX-2", origin_file_pattern="ltx-2-19b-dev.safetensors", **vram_config),
ModelConfig(model_id="DiffSynth-Studio/LTX-2-Repackage", origin_file_pattern="transformer.safetensors", **vram_config),
ModelConfig(model_id="DiffSynth-Studio/LTX-2-Repackage", origin_file_pattern="text_encoder_post_modules.safetensors", **vram_config),
ModelConfig(model_id="DiffSynth-Studio/LTX-2-Repackage", origin_file_pattern="video_vae_decoder.safetensors", **vram_config),
ModelConfig(model_id="DiffSynth-Studio/LTX-2-Repackage", origin_file_pattern="audio_vae_decoder.safetensors", **vram_config),
ModelConfig(model_id="DiffSynth-Studio/LTX-2-Repackage", origin_file_pattern="audio_vocoder.safetensors", **vram_config),
ModelConfig(model_id="DiffSynth-Studio/LTX-2-Repackage", origin_file_pattern="video_vae_encoder.safetensors", **vram_config),
],
tokenizer_config=ModelConfig(model_id="google/gemma-3-12b-it-qat-q4_0-unquantized"),
vram_limit=torch.cuda.mem_get_info("cuda")[1] / (1024 ** 3) - 0.5,

7
examples/ltx2/model_inference_low_vram/LTX-2-I2AV-TwoStage.py
View File

@@ -19,7 +19,12 @@ pipe = LTX2AudioVideoPipeline.from_pretrained(
device="cuda",
model_configs=[
ModelConfig(model_id="google/gemma-3-12b-it-qat-q4_0-unquantized", origin_file_pattern="model-*.safetensors", **vram_config),
ModelConfig(model_id="Lightricks/LTX-2", origin_file_pattern="ltx-2-19b-dev.safetensors", **vram_config),
ModelConfig(model_id="DiffSynth-Studio/LTX-2-Repackage", origin_file_pattern="transformer.safetensors", **vram_config),
ModelConfig(model_id="DiffSynth-Studio/LTX-2-Repackage", origin_file_pattern="text_encoder_post_modules.safetensors", **vram_config),
ModelConfig(model_id="DiffSynth-Studio/LTX-2-Repackage", origin_file_pattern="video_vae_decoder.safetensors", **vram_config),
ModelConfig(model_id="DiffSynth-Studio/LTX-2-Repackage", origin_file_pattern="audio_vae_decoder.safetensors", **vram_config),
ModelConfig(model_id="DiffSynth-Studio/LTX-2-Repackage", origin_file_pattern="audio_vocoder.safetensors", **vram_config),
ModelConfig(model_id="DiffSynth-Studio/LTX-2-Repackage", origin_file_pattern="video_vae_encoder.safetensors", **vram_config),
ModelConfig(model_id="Lightricks/LTX-2", origin_file_pattern="ltx-2-spatial-upscaler-x2-1.0.safetensors", **vram_config),
],
tokenizer_config=ModelConfig(model_id="google/gemma-3-12b-it-qat-q4_0-unquantized"),

7
examples/ltx2/model_inference_low_vram/LTX-2-T2AV-Camera-Control-Dolly-In.py
View File

@@ -17,7 +17,12 @@ pipe = LTX2AudioVideoPipeline.from_pretrained(
device="cuda",
model_configs=[
ModelConfig(model_id="google/gemma-3-12b-it-qat-q4_0-unquantized", origin_file_pattern="model-*.safetensors", **vram_config),
ModelConfig(model_id="Lightricks/LTX-2", origin_file_pattern="ltx-2-19b-dev.safetensors", **vram_config),
ModelConfig(model_id="DiffSynth-Studio/LTX-2-Repackage", origin_file_pattern="transformer.safetensors", **vram_config),
ModelConfig(model_id="DiffSynth-Studio/LTX-2-Repackage", origin_file_pattern="text_encoder_post_modules.safetensors", **vram_config),
ModelConfig(model_id="DiffSynth-Studio/LTX-2-Repackage", origin_file_pattern="video_vae_decoder.safetensors", **vram_config),
ModelConfig(model_id="DiffSynth-Studio/LTX-2-Repackage", origin_file_pattern="audio_vae_decoder.safetensors", **vram_config),
ModelConfig(model_id="DiffSynth-Studio/LTX-2-Repackage", origin_file_pattern="audio_vocoder.safetensors", **vram_config),
ModelConfig(model_id="DiffSynth-Studio/LTX-2-Repackage", origin_file_pattern="video_vae_encoder.safetensors", **vram_config),
ModelConfig(model_id="Lightricks/LTX-2", origin_file_pattern="ltx-2-spatial-upscaler-x2-1.0.safetensors", **vram_config),
],
tokenizer_config=ModelConfig(model_id="google/gemma-3-12b-it-qat-q4_0-unquantized"),

7
examples/ltx2/model_inference_low_vram/LTX-2-T2AV-Camera-Control-Dolly-Left.py
View File

@@ -17,7 +17,12 @@ pipe = LTX2AudioVideoPipeline.from_pretrained(
device="cuda",
model_configs=[
ModelConfig(model_id="google/gemma-3-12b-it-qat-q4_0-unquantized", origin_file_pattern="model-*.safetensors", **vram_config),
ModelConfig(model_id="Lightricks/LTX-2", origin_file_pattern="ltx-2-19b-dev.safetensors", **vram_config),
ModelConfig(model_id="DiffSynth-Studio/LTX-2-Repackage", origin_file_pattern="transformer.safetensors", **vram_config),
ModelConfig(model_id="DiffSynth-Studio/LTX-2-Repackage", origin_file_pattern="text_encoder_post_modules.safetensors", **vram_config),
ModelConfig(model_id="DiffSynth-Studio/LTX-2-Repackage", origin_file_pattern="video_vae_decoder.safetensors", **vram_config),
ModelConfig(model_id="DiffSynth-Studio/LTX-2-Repackage", origin_file_pattern="audio_vae_decoder.safetensors", **vram_config),
ModelConfig(model_id="DiffSynth-Studio/LTX-2-Repackage", origin_file_pattern="audio_vocoder.safetensors", **vram_config),
ModelConfig(model_id="DiffSynth-Studio/LTX-2-Repackage", origin_file_pattern="video_vae_encoder.safetensors", **vram_config),
ModelConfig(model_id="Lightricks/LTX-2", origin_file_pattern="ltx-2-spatial-upscaler-x2-1.0.safetensors", **vram_config),
],
tokenizer_config=ModelConfig(model_id="google/gemma-3-12b-it-qat-q4_0-unquantized"),

7
examples/ltx2/model_inference_low_vram/LTX-2-T2AV-Camera-Control-Dolly-Out.py
View File

@@ -17,7 +17,12 @@ pipe = LTX2AudioVideoPipeline.from_pretrained(
device="cuda",
model_configs=[
ModelConfig(model_id="google/gemma-3-12b-it-qat-q4_0-unquantized", origin_file_pattern="model-*.safetensors", **vram_config),
ModelConfig(model_id="Lightricks/LTX-2", origin_file_pattern="ltx-2-19b-dev.safetensors", **vram_config),
ModelConfig(model_id="DiffSynth-Studio/LTX-2-Repackage", origin_file_pattern="transformer.safetensors", **vram_config),
ModelConfig(model_id="DiffSynth-Studio/LTX-2-Repackage", origin_file_pattern="text_encoder_post_modules.safetensors", **vram_config),
ModelConfig(model_id="DiffSynth-Studio/LTX-2-Repackage", origin_file_pattern="video_vae_decoder.safetensors", **vram_config),
ModelConfig(model_id="DiffSynth-Studio/LTX-2-Repackage", origin_file_pattern="audio_vae_decoder.safetensors", **vram_config),
ModelConfig(model_id="DiffSynth-Studio/LTX-2-Repackage", origin_file_pattern="audio_vocoder.safetensors", **vram_config),
ModelConfig(model_id="DiffSynth-Studio/LTX-2-Repackage", origin_file_pattern="video_vae_encoder.safetensors", **vram_config),
ModelConfig(model_id="Lightricks/LTX-2", origin_file_pattern="ltx-2-spatial-upscaler-x2-1.0.safetensors", **vram_config),
],
tokenizer_config=ModelConfig(model_id="google/gemma-3-12b-it-qat-q4_0-unquantized"),

7
examples/ltx2/model_inference_low_vram/LTX-2-T2AV-Camera-Control-Dolly-Right.py
View File

@@ -17,7 +17,12 @@ pipe = LTX2AudioVideoPipeline.from_pretrained(
device="cuda",
model_configs=[
ModelConfig(model_id="google/gemma-3-12b-it-qat-q4_0-unquantized", origin_file_pattern="model-*.safetensors", **vram_config),
ModelConfig(model_id="Lightricks/LTX-2", origin_file_pattern="ltx-2-19b-dev.safetensors", **vram_config),
ModelConfig(model_id="DiffSynth-Studio/LTX-2-Repackage", origin_file_pattern="transformer.safetensors", **vram_config),
ModelConfig(model_id="DiffSynth-Studio/LTX-2-Repackage", origin_file_pattern="text_encoder_post_modules.safetensors", **vram_config),
ModelConfig(model_id="DiffSynth-Studio/LTX-2-Repackage", origin_file_pattern="video_vae_decoder.safetensors", **vram_config),
ModelConfig(model_id="DiffSynth-Studio/LTX-2-Repackage", origin_file_pattern="audio_vae_decoder.safetensors", **vram_config),
ModelConfig(model_id="DiffSynth-Studio/LTX-2-Repackage", origin_file_pattern="audio_vocoder.safetensors", **vram_config),
ModelConfig(model_id="DiffSynth-Studio/LTX-2-Repackage", origin_file_pattern="video_vae_encoder.safetensors", **vram_config),
ModelConfig(model_id="Lightricks/LTX-2", origin_file_pattern="ltx-2-spatial-upscaler-x2-1.0.safetensors", **vram_config),
],
tokenizer_config=ModelConfig(model_id="google/gemma-3-12b-it-qat-q4_0-unquantized"),

7
examples/ltx2/model_inference_low_vram/LTX-2-T2AV-Camera-Control-Jib-Down.py
View File

@@ -17,7 +17,12 @@ pipe = LTX2AudioVideoPipeline.from_pretrained(
device="cuda",
model_configs=[
ModelConfig(model_id="google/gemma-3-12b-it-qat-q4_0-unquantized", origin_file_pattern="model-*.safetensors", **vram_config),
ModelConfig(model_id="Lightricks/LTX-2", origin_file_pattern="ltx-2-19b-dev.safetensors", **vram_config),
ModelConfig(model_id="DiffSynth-Studio/LTX-2-Repackage", origin_file_pattern="transformer.safetensors", **vram_config),
ModelConfig(model_id="DiffSynth-Studio/LTX-2-Repackage", origin_file_pattern="text_encoder_post_modules.safetensors", **vram_config),
ModelConfig(model_id="DiffSynth-Studio/LTX-2-Repackage", origin_file_pattern="video_vae_decoder.safetensors", **vram_config),
ModelConfig(model_id="DiffSynth-Studio/LTX-2-Repackage", origin_file_pattern="audio_vae_decoder.safetensors", **vram_config),
ModelConfig(model_id="DiffSynth-Studio/LTX-2-Repackage", origin_file_pattern="audio_vocoder.safetensors", **vram_config),
ModelConfig(model_id="DiffSynth-Studio/LTX-2-Repackage", origin_file_pattern="video_vae_encoder.safetensors", **vram_config),
ModelConfig(model_id="Lightricks/LTX-2", origin_file_pattern="ltx-2-spatial-upscaler-x2-1.0.safetensors", **vram_config),
],
tokenizer_config=ModelConfig(model_id="google/gemma-3-12b-it-qat-q4_0-unquantized"),

7
examples/ltx2/model_inference_low_vram/LTX-2-T2AV-Camera-Control-Jib-Up.py
View File

@@ -17,7 +17,12 @@ pipe = LTX2AudioVideoPipeline.from_pretrained(
device="cuda",
model_configs=[
ModelConfig(model_id="google/gemma-3-12b-it-qat-q4_0-unquantized", origin_file_pattern="model-*.safetensors", **vram_config),
ModelConfig(model_id="Lightricks/LTX-2", origin_file_pattern="ltx-2-19b-dev.safetensors", **vram_config),
ModelConfig(model_id="DiffSynth-Studio/LTX-2-Repackage", origin_file_pattern="transformer.safetensors", **vram_config),
ModelConfig(model_id="DiffSynth-Studio/LTX-2-Repackage", origin_file_pattern="text_encoder_post_modules.safetensors", **vram_config),
ModelConfig(model_id="DiffSynth-Studio/LTX-2-Repackage", origin_file_pattern="video_vae_decoder.safetensors", **vram_config),
ModelConfig(model_id="DiffSynth-Studio/LTX-2-Repackage", origin_file_pattern="audio_vae_decoder.safetensors", **vram_config),
ModelConfig(model_id="DiffSynth-Studio/LTX-2-Repackage", origin_file_pattern="audio_vocoder.safetensors", **vram_config),
ModelConfig(model_id="DiffSynth-Studio/LTX-2-Repackage", origin_file_pattern="video_vae_encoder.safetensors", **vram_config),
ModelConfig(model_id="Lightricks/LTX-2", origin_file_pattern="ltx-2-spatial-upscaler-x2-1.0.safetensors", **vram_config),
],
tokenizer_config=ModelConfig(model_id="google/gemma-3-12b-it-qat-q4_0-unquantized"),

7
examples/ltx2/model_inference_low_vram/LTX-2-T2AV-Camera-Control-Static.py
View File

@@ -17,7 +17,12 @@ pipe = LTX2AudioVideoPipeline.from_pretrained(
device="cuda",
model_configs=[
ModelConfig(model_id="google/gemma-3-12b-it-qat-q4_0-unquantized", origin_file_pattern="model-*.safetensors", **vram_config),
ModelConfig(model_id="Lightricks/LTX-2", origin_file_pattern="ltx-2-19b-dev.safetensors", **vram_config),
ModelConfig(model_id="DiffSynth-Studio/LTX-2-Repackage", origin_file_pattern="transformer.safetensors", **vram_config),
ModelConfig(model_id="DiffSynth-Studio/LTX-2-Repackage", origin_file_pattern="text_encoder_post_modules.safetensors", **vram_config),
ModelConfig(model_id="DiffSynth-Studio/LTX-2-Repackage", origin_file_pattern="video_vae_decoder.safetensors", **vram_config),
ModelConfig(model_id="DiffSynth-Studio/LTX-2-Repackage", origin_file_pattern="audio_vae_decoder.safetensors", **vram_config),
ModelConfig(model_id="DiffSynth-Studio/LTX-2-Repackage", origin_file_pattern="audio_vocoder.safetensors", **vram_config),
ModelConfig(model_id="DiffSynth-Studio/LTX-2-Repackage", origin_file_pattern="video_vae_encoder.safetensors", **vram_config),
ModelConfig(model_id="Lightricks/LTX-2", origin_file_pattern="ltx-2-spatial-upscaler-x2-1.0.safetensors", **vram_config),
],
tokenizer_config=ModelConfig(model_id="google/gemma-3-12b-it-qat-q4_0-unquantized"),

7
examples/ltx2/model_inference_low_vram/LTX-2-T2AV-DistilledPipeline.py
View File

@@ -17,7 +17,12 @@ pipe = LTX2AudioVideoPipeline.from_pretrained(
device="cuda",
model_configs=[
ModelConfig(model_id="google/gemma-3-12b-it-qat-q4_0-unquantized", origin_file_pattern="model-*.safetensors", **vram_config),
ModelConfig(model_id="Lightricks/LTX-2", origin_file_pattern="ltx-2-19b-distilled.safetensors", **vram_config),
ModelConfig(model_id="DiffSynth-Studio/LTX-2-Repackage", origin_file_pattern="transformer_distilled.safetensors", **vram_config),
ModelConfig(model_id="DiffSynth-Studio/LTX-2-Repackage", origin_file_pattern="text_encoder_post_modules.safetensors", **vram_config),
ModelConfig(model_id="DiffSynth-Studio/LTX-2-Repackage", origin_file_pattern="video_vae_decoder.safetensors", **vram_config),
ModelConfig(model_id="DiffSynth-Studio/LTX-2-Repackage", origin_file_pattern="audio_vae_decoder.safetensors", **vram_config),
ModelConfig(model_id="DiffSynth-Studio/LTX-2-Repackage", origin_file_pattern="audio_vocoder.safetensors", **vram_config),
ModelConfig(model_id="DiffSynth-Studio/LTX-2-Repackage", origin_file_pattern="video_vae_encoder.safetensors", **vram_config),
ModelConfig(model_id="Lightricks/LTX-2", origin_file_pattern="ltx-2-spatial-upscaler-x2-1.0.safetensors", **vram_config),
],
tokenizer_config=ModelConfig(model_id="google/gemma-3-12b-it-qat-q4_0-unquantized"),

77
examples/ltx2/model_inference_low_vram/LTX-2-T2AV-IC-LoRA-Detailer.py Normal file
View File

@@ -0,0 +1,77 @@
import torch
from diffsynth.pipelines.ltx2_audio_video import LTX2AudioVideoPipeline, ModelConfig
from diffsynth.utils.data.media_io_ltx2 import write_video_audio_ltx2
from diffsynth.utils.data import VideoData
from modelscope import dataset_snapshot_download
vram_config = {
"offload_dtype": torch.float8_e5m2,
"offload_device": "cpu",
"onload_dtype": torch.float8_e5m2,
"onload_device": "cpu",
"preparing_dtype": torch.float8_e5m2,
"preparing_device": "cuda",
"computation_dtype": torch.bfloat16,
"computation_device": "cuda",
}
pipe = LTX2AudioVideoPipeline.from_pretrained(
torch_dtype=torch.bfloat16,
device="cuda",
model_configs=[
ModelConfig(model_id="google/gemma-3-12b-it-qat-q4_0-unquantized", origin_file_pattern="model-*.safetensors", **vram_config),
ModelConfig(model_id="DiffSynth-Studio/LTX-2-Repackage", origin_file_pattern="transformer.safetensors", **vram_config),
ModelConfig(model_id="DiffSynth-Studio/LTX-2-Repackage", origin_file_pattern="text_encoder_post_modules.safetensors", **vram_config),
ModelConfig(model_id="DiffSynth-Studio/LTX-2-Repackage", origin_file_pattern="video_vae_decoder.safetensors", **vram_config),
ModelConfig(model_id="DiffSynth-Studio/LTX-2-Repackage", origin_file_pattern="audio_vae_decoder.safetensors", **vram_config),
ModelConfig(model_id="DiffSynth-Studio/LTX-2-Repackage", origin_file_pattern="audio_vocoder.safetensors", **vram_config),
ModelConfig(model_id="DiffSynth-Studio/LTX-2-Repackage", origin_file_pattern="video_vae_encoder.safetensors", **vram_config),
ModelConfig(model_id="Lightricks/LTX-2", origin_file_pattern="ltx-2-spatial-upscaler-x2-1.0.safetensors", **vram_config),
],
tokenizer_config=ModelConfig(model_id="google/gemma-3-12b-it-qat-q4_0-unquantized"),
stage2_lora_config=ModelConfig(model_id="Lightricks/LTX-2", origin_file_pattern="ltx-2-19b-distilled-lora-384.safetensors"),
)
pipe.load_lora(pipe.dit, ModelConfig(model_id="Lightricks/LTX-2-19b-IC-LoRA-Detailer", origin_file_pattern="ltx-2-19b-ic-lora-detailer.safetensors"))
dataset_snapshot_download("DiffSynth-Studio/example_video_dataset", allow_file_pattern="ltx2/*", local_dir="data/example_video_dataset")
prompt = "[VISUAL]:Two cute orange cats, wearing boxing gloves, stand on a boxing ring and fight each other. [SOUNDS]:the sound of two cats boxing"
negative_prompt = (
"blurry, out of focus, overexposed, underexposed, low contrast, washed out colors, excessive noise, "
"grainy texture, poor lighting, flickering, motion blur, distorted proportions, unnatural skin tones, "
"deformed facial features, asymmetrical face, missing facial features, extra limbs, disfigured hands, "
"wrong hand count, artifacts around text, inconsistent perspective, camera shake, incorrect depth of "
"field, background too sharp, background clutter, distracting reflections, harsh shadows, inconsistent "
"lighting direction, color banding, cartoonish rendering, 3D CGI look, unrealistic materials, uncanny "
"valley effect, incorrect ethnicity, wrong gender, exaggerated expressions, wrong gaze direction, "
"mismatched lip sync, silent or muted audio, distorted voice, robotic voice, echo, background noise, "
"off-sync audio, incorrect dialogue, added dialogue, repetitive speech, jittery movement, awkward "
"pauses, incorrect timing, unnatural transitions, inconsistent framing, tilted camera, flat lighting, "
"inconsistent tone, cinematic oversaturation, stylized filters, or AI artifacts."
)
height, width, num_frames = 512 * 2, 768 * 2, 121
ref_scale_factor = 1
frame_rate = 24
# the frame rate of the video should better be the same with the reference video
# the spatial resolution of the first frame should be the resolution of stage 1 video generation divided by ref_scale_factor
input_video = VideoData("data/example_video_dataset/ltx2/video1.mp4", height=height // ref_scale_factor // 2, width=width // ref_scale_factor // 2)
input_video = input_video.raw_data()
video, audio = pipe(
prompt=prompt,
negative_prompt=negative_prompt,
seed=43,
height=height,
width=width,
num_frames=num_frames,
frame_rate=frame_rate,
in_context_videos=[input_video],
in_context_downsample_factor=ref_scale_factor,
tiled=True,
use_two_stage_pipeline=True,
clear_lora_before_state_two=True,
)
write_video_audio_ltx2(
video=video,
audio=audio,
output_path='ltx2_twostage_iclora.mp4',
fps=frame_rate,
audio_sample_rate=24000,
)

77
examples/ltx2/model_inference_low_vram/LTX-2-T2AV-IC-LoRA-Union-Control.py Normal file
View File

@@ -0,0 +1,77 @@
import torch
from diffsynth.pipelines.ltx2_audio_video import LTX2AudioVideoPipeline, ModelConfig
from diffsynth.utils.data.media_io_ltx2 import write_video_audio_ltx2
from diffsynth.utils.data import VideoData
from modelscope import dataset_snapshot_download
vram_config = {
"offload_dtype": torch.float8_e5m2,
"offload_device": "cpu",
"onload_dtype": torch.float8_e5m2,
"onload_device": "cpu",
"preparing_dtype": torch.float8_e5m2,
"preparing_device": "cuda",
"computation_dtype": torch.bfloat16,
"computation_device": "cuda",
}
pipe = LTX2AudioVideoPipeline.from_pretrained(
torch_dtype=torch.bfloat16,
device="cuda",
model_configs=[
ModelConfig(model_id="google/gemma-3-12b-it-qat-q4_0-unquantized", origin_file_pattern="model-*.safetensors", **vram_config),
ModelConfig(model_id="DiffSynth-Studio/LTX-2-Repackage", origin_file_pattern="transformer.safetensors", **vram_config),
ModelConfig(model_id="DiffSynth-Studio/LTX-2-Repackage", origin_file_pattern="text_encoder_post_modules.safetensors", **vram_config),
ModelConfig(model_id="DiffSynth-Studio/LTX-2-Repackage", origin_file_pattern="video_vae_decoder.safetensors", **vram_config),
ModelConfig(model_id="DiffSynth-Studio/LTX-2-Repackage", origin_file_pattern="audio_vae_decoder.safetensors", **vram_config),
ModelConfig(model_id="DiffSynth-Studio/LTX-2-Repackage", origin_file_pattern="audio_vocoder.safetensors", **vram_config),
ModelConfig(model_id="DiffSynth-Studio/LTX-2-Repackage", origin_file_pattern="video_vae_encoder.safetensors", **vram_config),
ModelConfig(model_id="Lightricks/LTX-2", origin_file_pattern="ltx-2-spatial-upscaler-x2-1.0.safetensors", **vram_config),
],
tokenizer_config=ModelConfig(model_id="google/gemma-3-12b-it-qat-q4_0-unquantized"),
stage2_lora_config=ModelConfig(model_id="Lightricks/LTX-2", origin_file_pattern="ltx-2-19b-distilled-lora-384.safetensors"),
)
pipe.load_lora(pipe.dit, ModelConfig(model_id="Lightricks/LTX-2-19b-IC-LoRA-Union-Control", origin_file_pattern="ltx-2-19b-ic-lora-union-control-ref0.5.safetensors"))
dataset_snapshot_download("DiffSynth-Studio/example_video_dataset", allow_file_pattern="ltx2/*", local_dir="data/example_video_dataset")
prompt = "[VISUAL]:Two cute orange cats, wearing boxing gloves, stand on a boxing ring and fight each other. [SOUNDS]:the sound of two cats boxing"
negative_prompt = (
"blurry, out of focus, overexposed, underexposed, low contrast, washed out colors, excessive noise, "
"grainy texture, poor lighting, flickering, motion blur, distorted proportions, unnatural skin tones, "
"deformed facial features, asymmetrical face, missing facial features, extra limbs, disfigured hands, "
"wrong hand count, artifacts around text, inconsistent perspective, camera shake, incorrect depth of "
"field, background too sharp, background clutter, distracting reflections, harsh shadows, inconsistent "
"lighting direction, color banding, cartoonish rendering, 3D CGI look, unrealistic materials, uncanny "
"valley effect, incorrect ethnicity, wrong gender, exaggerated expressions, wrong gaze direction, "
"mismatched lip sync, silent or muted audio, distorted voice, robotic voice, echo, background noise, "
"off-sync audio, incorrect dialogue, added dialogue, repetitive speech, jittery movement, awkward "
"pauses, incorrect timing, unnatural transitions, inconsistent framing, tilted camera, flat lighting, "
"inconsistent tone, cinematic oversaturation, stylized filters, or AI artifacts."
)
height, width, num_frames = 512 * 2, 768 * 2, 121
ref_scale_factor = 2
frame_rate = 24
# the frame rate of the video should better be the same with the reference video
# the spatial resolution of the first frame should be the resolution of stage 1 video generation divided by ref_scale_factor
input_video = VideoData("data/example_video_dataset/ltx2/depth_video.mp4", height=height // ref_scale_factor // 2, width=width // ref_scale_factor // 2)
input_video = input_video.raw_data()
video, audio = pipe(
prompt=prompt,
negative_prompt=negative_prompt,
seed=43,
height=height,
width=width,
num_frames=num_frames,
frame_rate=frame_rate,
in_context_videos=[input_video],
in_context_downsample_factor=ref_scale_factor,
tiled=True,
use_two_stage_pipeline=True,
clear_lora_before_state_two=True,
)
write_video_audio_ltx2(
video=video,
audio=audio,
output_path='ltx2_twostage_iclora.mp4',
fps=frame_rate,
audio_sample_rate=24000,
)

26
examples/ltx2/model_inference_low_vram/LTX-2-T2AV-OneStage.py
View File

@@ -12,16 +12,40 @@ vram_config = {
"computation_dtype": torch.bfloat16,
"computation_device": "cuda",
}
"""
Offical model repo: https://www.modelscope.cn/models/Lightricks/LTX-2
Repackaged model repo: https://www.modelscope.cn/models/DiffSynth-Studio/LTX-2-Repackage
For base models of LTX-2, offical checkpoint (with model config ModelConfig(model_id="Lightricks/LTX-2", origin_file_pattern="ltx-2-19b-dev.safetensors"))
and repackaged checkpoints (with model config ModelConfig(model_id="DiffSynth-Studio/LTX-2-Repackage", origin_file_pattern="*.safetensors")) are both supported.
We have repackeged the official checkpoints in DiffSynth-Studio/LTX-2-Repackage repo to support separate loading of different submodules,
and avoid redundant memory usage when users only want to use part of the model.
"""
# use the repackaged modelconfig from "DiffSynth-Studio/LTX-2-Repackage" to avoid redundant model loading
pipe = LTX2AudioVideoPipeline.from_pretrained(
torch_dtype=torch.bfloat16,
device="cuda",
model_configs=[
ModelConfig(model_id="google/gemma-3-12b-it-qat-q4_0-unquantized", origin_file_pattern="model-*.safetensors", **vram_config),
ModelConfig(model_id="Lightricks/LTX-2", origin_file_pattern="ltx-2-19b-dev.safetensors", **vram_config),
ModelConfig(model_id="DiffSynth-Studio/LTX-2-Repackage", origin_file_pattern="transformer.safetensors", **vram_config),
ModelConfig(model_id="DiffSynth-Studio/LTX-2-Repackage", origin_file_pattern="text_encoder_post_modules.safetensors", **vram_config),
ModelConfig(model_id="DiffSynth-Studio/LTX-2-Repackage", origin_file_pattern="video_vae_decoder.safetensors", **vram_config),
ModelConfig(model_id="DiffSynth-Studio/LTX-2-Repackage", origin_file_pattern="audio_vae_decoder.safetensors", **vram_config),
ModelConfig(model_id="DiffSynth-Studio/LTX-2-Repackage", origin_file_pattern="audio_vocoder.safetensors", **vram_config),
],
tokenizer_config=ModelConfig(model_id="google/gemma-3-12b-it-qat-q4_0-unquantized"),
vram_limit=torch.cuda.mem_get_info("cuda")[1] / (1024 ** 3) - 0.5,
)
# use the following modelconfig if you want to initialize model from offical checkpoints from "Lightricks/LTX-2"
# pipe = LTX2AudioVideoPipeline.from_pretrained(
# torch_dtype=torch.bfloat16,
# device="cuda",
# model_configs=[
# ModelConfig(model_id="google/gemma-3-12b-it-qat-q4_0-unquantized", origin_file_pattern="model-*.safetensors", **vram_config),
# ModelConfig(model_id="Lightricks/LTX-2", origin_file_pattern="ltx-2-19b-dev.safetensors", **vram_config),
# ],
# tokenizer_config=ModelConfig(model_id="google/gemma-3-12b-it-qat-q4_0-unquantized"),
# vram_limit=torch.cuda.mem_get_info("cuda")[1] / (1024 ** 3) - 0.5,
# )
prompt = "A girl is very happy, she is speaking: “I enjoy working with Diffsynth-Studio, it's a perfect framework.”"
negative_prompt = "blurry, out of focus, overexposed, underexposed, low contrast, washed out colors, excessive noise, grainy texture, poor lighting, flickering, motion blur, distorted proportions, unnatural skin tones, deformed facial features, asymmetrical face, missing facial features, extra limbs, disfigured hands, wrong hand count, artifacts around text, inconsistent perspective, camera shake, incorrect depth of field, background too sharp, background clutter, distracting reflections, harsh shadows, inconsistent lighting direction, color banding, cartoonish rendering, 3D CGI look, unrealistic materials, uncanny valley effect, incorrect ethnicity, wrong gender, exaggerated expressions, wrong gaze direction, mismatched lip sync, silent or muted audio, distorted voice, robotic voice, echo, background noise, off-sync audio, incorrect dialogue, added dialogue, repetitive speech, jittery movement, awkward pauses, incorrect timing, unnatural transitions, inconsistent framing, tilted camera, flat lighting, inconsistent tone, cinematic oversaturation, stylized filters, or AI artifacts."
height, width, num_frames = 512, 768, 121

29
examples/ltx2/model_inference_low_vram/LTX-2-T2AV-TwoStage.py
View File

@@ -12,18 +12,45 @@ vram_config = {
"computation_dtype": torch.bfloat16,
"computation_device": "cuda",
}
"""
Offical model repo: https://www.modelscope.cn/models/Lightricks/LTX-2
Repackaged model repo: https://www.modelscope.cn/models/DiffSynth-Studio/LTX-2-Repackage
For base models of LTX-2, offical checkpoint (with model config ModelConfig(model_id="Lightricks/LTX-2", origin_file_pattern="ltx-2-19b-dev.safetensors"))
and repackaged checkpoints (with model config ModelConfig(model_id="DiffSynth-Studio/LTX-2-Repackage", origin_file_pattern="*.safetensors")) are both supported.
We have repackeged the official checkpoints in DiffSynth-Studio/LTX-2-Repackage repo to support separate loading of different submodules,
and avoid redundant memory usage when users only want to use part of the model.
"""
# use the repackaged modelconfig from "DiffSynth-Studio/LTX-2-Repackage" to avoid redundant model loading
pipe = LTX2AudioVideoPipeline.from_pretrained(
torch_dtype=torch.bfloat16,
device="cuda",
model_configs=[
ModelConfig(model_id="google/gemma-3-12b-it-qat-q4_0-unquantized", origin_file_pattern="model-*.safetensors", **vram_config),
ModelConfig(model_id="Lightricks/LTX-2", origin_file_pattern="ltx-2-19b-dev.safetensors", **vram_config),
ModelConfig(model_id="DiffSynth-Studio/LTX-2-Repackage", origin_file_pattern="transformer.safetensors", **vram_config),
ModelConfig(model_id="DiffSynth-Studio/LTX-2-Repackage", origin_file_pattern="text_encoder_post_modules.safetensors", **vram_config),
ModelConfig(model_id="DiffSynth-Studio/LTX-2-Repackage", origin_file_pattern="video_vae_decoder.safetensors", **vram_config),
ModelConfig(model_id="DiffSynth-Studio/LTX-2-Repackage", origin_file_pattern="audio_vae_decoder.safetensors", **vram_config),
ModelConfig(model_id="DiffSynth-Studio/LTX-2-Repackage", origin_file_pattern="audio_vocoder.safetensors", **vram_config),
ModelConfig(model_id="DiffSynth-Studio/LTX-2-Repackage", origin_file_pattern="video_vae_encoder.safetensors", **vram_config),
ModelConfig(model_id="Lightricks/LTX-2", origin_file_pattern="ltx-2-spatial-upscaler-x2-1.0.safetensors", **vram_config),
],
tokenizer_config=ModelConfig(model_id="google/gemma-3-12b-it-qat-q4_0-unquantized"),
stage2_lora_config=ModelConfig(model_id="Lightricks/LTX-2", origin_file_pattern="ltx-2-19b-distilled-lora-384.safetensors"),
vram_limit=torch.cuda.mem_get_info("cuda")[1] / (1024 ** 3) - 0.5,
)
# use the following modelconfig if you want to initialize model from offical checkpoints from "Lightricks/LTX-2"
# pipe = LTX2AudioVideoPipeline.from_pretrained(
# torch_dtype=torch.bfloat16,
# device="cuda",
# model_configs=[
# ModelConfig(model_id="google/gemma-3-12b-it-qat-q4_0-unquantized", origin_file_pattern="model-*.safetensors", **vram_config),
# ModelConfig(model_id="Lightricks/LTX-2", origin_file_pattern="ltx-2-19b-dev.safetensors", **vram_config),
# ModelConfig(model_id="Lightricks/LTX-2", origin_file_pattern="ltx-2-spatial-upscaler-x2-1.0.safetensors", **vram_config),
# ],
# tokenizer_config=ModelConfig(model_id="google/gemma-3-12b-it-qat-q4_0-unquantized"),
# stage2_lora_config=ModelConfig(model_id="Lightricks/LTX-2", origin_file_pattern="ltx-2-19b-distilled-lora-384.safetensors"),
# vram_limit=torch.cuda.mem_get_info("cuda")[1] / (1024 ** 3) - 0.5,
# )
prompt = "A girl is very happy, she is speaking: “I enjoy working with Diffsynth-Studio, it's a perfect framework.”"
negative_prompt = (

35
examples/ltx2/model_training/full/LTX-2-T2AV-splited.sh Normal file
View File

@@ -0,0 +1,35 @@
# Splited Training
accelerate launch examples/ltx2/model_training/train.py \
--dataset_base_path data/example_video_dataset/ltx2 \
--dataset_metadata_path data/example_video_dataset/ltx2_t2av.csv \
--data_file_keys "video,input_audio" \
--extra_inputs "input_audio" \
--height 512 \
--width 768 \
--num_frames 121 \
--dataset_repeat 1 \
--model_id_with_origin_paths "DiffSynth-Studio/LTX-2-Repackage:text_encoder_post_modules.safetensors,DiffSynth-Studio/LTX-2-Repackage:video_vae_encoder.safetensors,DiffSynth-Studio/LTX-2-Repackage:audio_vae_encoder.safetensors,google/gemma-3-12b-it-qat-q4_0-unquantized:model-*.safetensors" \
--learning_rate 1e-4 \
--num_epochs 5 \
--remove_prefix_in_ckpt "pipe.dit." \
--output_path "./models/train/LTX2-T2AV-full-splited-cache" \
--trainable_models "dit" \
--use_gradient_checkpointing \
--task "sft:data_process"
accelerate launch --config_file examples/qwen_image/model_training/full/accelerate_config_zero2offload.yaml examples/ltx2/model_training/train.py \
--dataset_base_path ./models/train/LTX2-T2AV-full-splited-cache \
--data_file_keys "video,input_audio" \
--extra_inputs "input_audio" \
--height 512 \
--width 768 \
--num_frames 121 \
--dataset_repeat 100 \
--model_id_with_origin_paths "DiffSynth-Studio/LTX-2-Repackage:transformer.safetensors" \
--learning_rate 1e-4 \
--num_epochs 5 \
--remove_prefix_in_ckpt "pipe.dit." \
--output_path "./models/train/LTX2-T2AV-full" \
--trainable_models "dit" \
--use_gradient_checkpointing \
--task "sft:train"

39
examples/ltx2/model_training/lora/LTX-2-T2AV-IC-LoRA-splited.sh Normal file
View File

@@ -0,0 +1,39 @@
# Splited Training
accelerate launch examples/ltx2/model_training/train.py \
--dataset_base_path data/example_video_dataset/ltx2 \
--dataset_metadata_path data/example_video_dataset/ltx2_t2av_iclora.json \
--data_file_keys "video,input_audio,in_context_videos" \
--extra_inputs "input_audio,in_context_videos,in_context_downsample_factor,frame_rate" \
--height 512 \
--width 768 \
--num_frames 81 \
--dataset_repeat 1 \
--model_id_with_origin_paths "DiffSynth-Studio/LTX-2-Repackage:text_encoder_post_modules.safetensors,DiffSynth-Studio/LTX-2-Repackage:video_vae_encoder.safetensors,DiffSynth-Studio/LTX-2-Repackage:audio_vae_encoder.safetensors,google/gemma-3-12b-it-qat-q4_0-unquantized:model-*.safetensors" \
--learning_rate 1e-4 \
--num_epochs 5 \
--remove_prefix_in_ckpt "pipe.dit." \
--output_path "./models/train/LTX2-T2AV-IC-LoRA-splited-cache" \
--lora_base_model "dit" \
--lora_target_modules "to_k,to_q,to_v,to_out.0" \
--lora_rank 32 \
--use_gradient_checkpointing \
--task "sft:data_process"
accelerate launch examples/ltx2/model_training/train.py \
--dataset_base_path ./models/train/LTX2-T2AV-IC-LoRA-splited-cache \
--data_file_keys "video,input_audio,in_context_videos" \
--extra_inputs "input_audio,in_context_videos,in_context_downsample_factor,frame_rate" \
--height 512 \
--width 768 \
--num_frames 81 \
--dataset_repeat 100 \
--model_id_with_origin_paths "DiffSynth-Studio/LTX-2-Repackage:transformer.safetensors" \
--learning_rate 1e-4 \
--num_epochs 5 \
--remove_prefix_in_ckpt "pipe.dit." \
--output_path "./models/train/LTX2-T2AV-IC-LoRA" \
--lora_base_model "dit" \
--lora_target_modules "to_k,to_q,to_v,to_out.0" \
--lora_rank 32 \
--use_gradient_checkpointing \
--task "sft:train"

56
examples/ltx2/model_training/lora/LTX-2-T2AV-noaudio.sh Normal file
View File

@@ -0,0 +1,56 @@
# single stage training
# accelerate launch examples/ltx2/model_training/train.py \
# --dataset_base_path data/example_video_dataset/ltx2 \
# --dataset_metadata_path data/example_video_dataset/ltx2_t2v.csv \
# --height 256 \
# --width 384 \
# --num_frames 25\
# --dataset_repeat 100 \
# --model_id_with_origin_paths "DiffSynth-Studio/LTX-2-Repackage:transformer.safetensors,DiffSynth-Studio/LTX-2-Repackage:text_encoder_post_modules.safetensors,DiffSynth-Studio/LTX-2-Repackage:video_vae_encoder.safetensors,DiffSynth-Studio/LTX-2-Repackage:audio_vae_encoder.safetensors,google/gemma-3-12b-it-qat-q4_0-unquantized:model-*.safetensors" \
# --learning_rate 1e-4 \
# --num_epochs 5 \
# --remove_prefix_in_ckpt "pipe.dit." \
# --output_path "./models/train/LTX2-T2AV-noaudio_lora" \
# --lora_base_model "dit" \
# --lora_target_modules "to_k,to_q,to_v,to_out.0" \
# --lora_rank 32 \
# --use_gradient_checkpointing \
# --find_unused_parameters
# Splited Training
accelerate launch examples/ltx2/model_training/train.py \
--dataset_base_path data/example_video_dataset/ltx2 \
--dataset_metadata_path data/example_video_dataset/ltx2_t2av.csv \
--height 512 \
--width 768 \
--num_frames 121\
--dataset_repeat 1 \
--model_id_with_origin_paths "DiffSynth-Studio/LTX-2-Repackage:text_encoder_post_modules.safetensors,DiffSynth-Studio/LTX-2-Repackage:video_vae_encoder.safetensors,DiffSynth-Studio/LTX-2-Repackage:audio_vae_encoder.safetensors,google/gemma-3-12b-it-qat-q4_0-unquantized:model-*.safetensors" \
--learning_rate 1e-4 \
--num_epochs 5 \
--remove_prefix_in_ckpt "pipe.dit." \
--output_path "./models/train/LTX2-T2AV-noaudio_lora-splited-cache" \
--lora_base_model "dit" \
--lora_target_modules "to_k,to_q,to_v,to_out.0" \
--lora_rank 32 \
--use_gradient_checkpointing \
--task "sft:data_process"
accelerate launch examples/ltx2/model_training/train.py \
--dataset_base_path ./models/train/LTX2-T2AV-noaudio_lora-splited-cache \
--height 512 \
--width 768 \
--num_frames 121\
--dataset_repeat 100 \
--model_id_with_origin_paths "DiffSynth-Studio/LTX-2-Repackage:transformer.safetensors" \
--learning_rate 1e-4 \
--num_epochs 5 \
--remove_prefix_in_ckpt "pipe.dit." \
--output_path "./models/train/LTX2-T2AV-noaudio_lora" \
--lora_base_model "dit" \
--lora_target_modules "to_k,to_q,to_v,to_out.0" \
--lora_rank 32 \
--use_gradient_checkpointing \
--task "sft:train"

60
examples/ltx2/model_training/lora/LTX-2-T2AV-splited.sh Normal file
View File

@@ -0,0 +1,60 @@
# Single Stage Training not recommended for T2AV due to the large memory consumption. Please use the Splited Training instead.
# accelerate launch examples/ltx2/model_training/train.py \
# --dataset_base_path data/example_video_dataset/ltx2 \
# --dataset_metadata_path data/example_video_dataset/ltx2_t2av.csv \
# --data_file_keys "video,input_audio" \
# --extra_inputs "input_audio" \
# --height 256 \
# --width 384 \
# --num_frames 25\
# --dataset_repeat 100 \
# --model_id_with_origin_paths "DiffSynth-Studio/LTX-2-Repackage:transformer.safetensors,DiffSynth-Studio/LTX-2-Repackage:text_encoder_post_modules.safetensors,DiffSynth-Studio/LTX-2-Repackage:video_vae_encoder.safetensors,DiffSynth-Studio/LTX-2-Repackage:audio_vae_encoder.safetensors,google/gemma-3-12b-it-qat-q4_0-unquantized:model-*.safetensors" \
# --learning_rate 1e-4 \
# --num_epochs 5 \
# --remove_prefix_in_ckpt "pipe.dit." \
# --output_path "./models/train/LTX2-T2AV_lora" \
# --lora_base_model "dit" \
# --lora_target_modules "to_k,to_q,to_v,to_out.0" \
# --lora_rank 32 \
# --use_gradient_checkpointing \
# --find_unused_parameters
# Splited Training
accelerate launch examples/ltx2/model_training/train.py \
--dataset_base_path data/example_video_dataset/ltx2 \
--dataset_metadata_path data/example_video_dataset/ltx2_t2av.csv \
--data_file_keys "video,input_audio" \
--extra_inputs "input_audio" \
--height 512 \
--width 768 \
--num_frames 121 \
--dataset_repeat 1 \
--model_id_with_origin_paths "DiffSynth-Studio/LTX-2-Repackage:text_encoder_post_modules.safetensors,DiffSynth-Studio/LTX-2-Repackage:video_vae_encoder.safetensors,DiffSynth-Studio/LTX-2-Repackage:audio_vae_encoder.safetensors,google/gemma-3-12b-it-qat-q4_0-unquantized:model-*.safetensors" \
--learning_rate 1e-4 \
--num_epochs 5 \
--remove_prefix_in_ckpt "pipe.dit." \
--output_path "./models/train/LTX2-T2AV_lora-splited-cache" \
--lora_base_model "dit" \
--lora_target_modules "to_k,to_q,to_v,to_out.0" \
--lora_rank 32 \
--use_gradient_checkpointing \
--task "sft:data_process"
accelerate launch examples/ltx2/model_training/train.py \
--dataset_base_path ./models/train/LTX2-T2AV_lora-splited-cache \
--data_file_keys "video,input_audio" \
--extra_inputs "input_audio" \
--height 512 \
--width 768 \
--num_frames 121 \
--dataset_repeat 100 \
--model_id_with_origin_paths "DiffSynth-Studio/LTX-2-Repackage:transformer.safetensors" \
--learning_rate 1e-4 \
--num_epochs 5 \
--remove_prefix_in_ckpt "pipe.dit." \
--output_path "./models/train/LTX2-T2AV_lora" \
--lora_base_model "dit" \
--lora_target_modules "to_k,to_q,to_v,to_out.0" \
--lora_rank 32 \
--use_gradient_checkpointing \
--task "sft:train"

104
examples/ltx2/model_training/scripts/split_model_statedicts.py Normal file
View File

@@ -0,0 +1,104 @@
from safetensors.torch import save_file
from diffsynth import hash_state_dict_keys
from diffsynth.core import load_state_dict
from diffsynth.models.model_loader import ModelPool
model_pool = ModelPool()
state_dict = load_state_dict("models/Lightricks/LTX-2/ltx-2-19b-dev.safetensors")
dit_state_dict = {}
for name in state_dict:
if name.startswith("model.diffusion_model."):
new_name = name.replace("model.diffusion_model.", "")
if new_name.startswith("audio_embeddings_connector.") or new_name.startswith("video_embeddings_connector."):
continue
dit_state_dict[name] = state_dict[name]
print(f"dit_state_dict keys hash: {hash_state_dict_keys(dit_state_dict)}")
save_file(dit_state_dict, "models/DiffSynth-Studio/LTX-2-Repackage/transformer.safetensors")
model_pool.auto_load_model(
"models/DiffSynth-Studio/LTX-2-Repackage/transformer.safetensors",
)
video_vae_encoder_state_dict = {}
for name in state_dict:
if name.startswith("vae.encoder."):
video_vae_encoder_state_dict[name] = state_dict[name]
elif name.startswith("vae.per_channel_statistics."):
video_vae_encoder_state_dict[name] = state_dict[name]
save_file(video_vae_encoder_state_dict, "models/DiffSynth-Studio/LTX-2-Repackage/video_vae_encoder.safetensors")
print(f"video_vae_encoder keys hash: {hash_state_dict_keys(video_vae_encoder_state_dict)}")
model_pool.auto_load_model("models/DiffSynth-Studio/LTX-2-Repackage/video_vae_encoder.safetensors")
video_vae_decoder_state_dict = {}
for name in state_dict:
if name.startswith("vae.decoder."):
video_vae_decoder_state_dict[name] = state_dict[name]
elif name.startswith("vae.per_channel_statistics."):
video_vae_decoder_state_dict[name] = state_dict[name]
save_file(video_vae_decoder_state_dict, "models/DiffSynth-Studio/LTX-2-Repackage/video_vae_decoder.safetensors")
print(f"video_vae_decoder keys hash: {hash_state_dict_keys(video_vae_decoder_state_dict)}")
model_pool.auto_load_model("models/DiffSynth-Studio/LTX-2-Repackage/video_vae_decoder.safetensors")
audio_vae_decoder_state_dict = {}
for name in state_dict:
if name.startswith("audio_vae.decoder."):
audio_vae_decoder_state_dict[name] = state_dict[name]
elif name.startswith("audio_vae.per_channel_statistics."):
audio_vae_decoder_state_dict[name] = state_dict[name]
save_file(audio_vae_decoder_state_dict, "models/DiffSynth-Studio/LTX-2-Repackage/audio_vae_decoder.safetensors")
print(f"audio_vae_decoder keys hash: {hash_state_dict_keys(audio_vae_decoder_state_dict)}")
model_pool.auto_load_model("models/DiffSynth-Studio/LTX-2-Repackage/audio_vae_decoder.safetensors")
audio_vae_encoder_state_dict = {}
for name in state_dict:
if name.startswith("audio_vae.encoder."):
audio_vae_encoder_state_dict[name] = state_dict[name]
elif name.startswith("audio_vae.per_channel_statistics."):
audio_vae_encoder_state_dict[name] = state_dict[name]
save_file(audio_vae_encoder_state_dict, "models/DiffSynth-Studio/LTX-2-Repackage/audio_vae_encoder.safetensors")
print(f"audio_vae_encoder keys hash: {hash_state_dict_keys(audio_vae_encoder_state_dict)}")
model_pool.auto_load_model("models/DiffSynth-Studio/LTX-2-Repackage/audio_vae_encoder.safetensors")
audio_vocoder_state_dict = {}
for name in state_dict:
if name.startswith("vocoder."):
audio_vocoder_state_dict[name] = state_dict[name]
save_file(audio_vocoder_state_dict, "models/DiffSynth-Studio/LTX-2-Repackage/audio_vocoder.safetensors")
print(f"audio_vocoder keys hash: {hash_state_dict_keys(audio_vocoder_state_dict)}")
model_pool.auto_load_model("models/DiffSynth-Studio/LTX-2-Repackage/audio_vocoder.safetensors")
text_encoder_post_modules_state_dict = {}
for name in state_dict:
if name.startswith("text_embedding_projection."):
text_encoder_post_modules_state_dict[name] = state_dict[name]
elif name.startswith("model.diffusion_model.video_embeddings_connector."):
text_encoder_post_modules_state_dict[name] = state_dict[name]
elif name.startswith("model.diffusion_model.audio_embeddings_connector."):
text_encoder_post_modules_state_dict[name] = state_dict[name]
save_file(text_encoder_post_modules_state_dict, "models/DiffSynth-Studio/LTX-2-Repackage/text_encoder_post_modules.safetensors")
print(f"text_encoder_post_modules keys hash: {hash_state_dict_keys(text_encoder_post_modules_state_dict)}")
model_pool.auto_load_model("models/DiffSynth-Studio/LTX-2-Repackage/text_encoder_post_modules.safetensors")
state_dict = load_state_dict("models/Lightricks/LTX-2/ltx-2-19b-distilled.safetensors")
dit_state_dict = {}
for name in state_dict:
if name.startswith("model.diffusion_model."):
new_name = name.replace("model.diffusion_model.", "")
if new_name.startswith("audio_embeddings_connector.") or new_name.startswith("video_embeddings_connector."):
continue
dit_state_dict[name] = state_dict[name]
print(f"dit_state_dict keys hash: {hash_state_dict_keys(dit_state_dict)}")
save_file(dit_state_dict, "models/DiffSynth-Studio/LTX-2-Repackage/transformer_distilled.safetensors")
model_pool.auto_load_model(
"models/DiffSynth-Studio/LTX-2-Repackage/transformer_distilled.safetensors",
)

167
examples/ltx2/model_training/train.py Normal file
View File

@@ -0,0 +1,167 @@
import torch, os, argparse, accelerate, warnings
from diffsynth.core import UnifiedDataset
from diffsynth.core.data.operators import LoadAudioWithTorchaudio, ToAbsolutePath, RouteByType, SequencialProcess
from diffsynth.pipelines.ltx2_audio_video import LTX2AudioVideoPipeline, ModelConfig
from diffsynth.diffusion import *
os.environ["TOKENIZERS_PARALLELISM"] = "false"
class LTX2TrainingModule(DiffusionTrainingModule):
def __init__(
self,
model_paths=None, model_id_with_origin_paths=None,
tokenizer_path=None,
trainable_models=None,
lora_base_model=None, lora_target_modules="", lora_rank=32, lora_checkpoint=None,
preset_lora_path=None, preset_lora_model=None,
use_gradient_checkpointing=True,
use_gradient_checkpointing_offload=False,
extra_inputs=None,
fp8_models=None,
offload_models=None,
device="cpu",
task="sft",
):
super().__init__()
# Warning
if not use_gradient_checkpointing:
warnings.warn("Gradient checkpointing is detected as disabled. To prevent out-of-memory errors, the training framework will forcibly enable gradient checkpointing.")
use_gradient_checkpointing = True
# Load models
model_configs = self.parse_model_configs(model_paths, model_id_with_origin_paths, fp8_models=fp8_models, offload_models=offload_models, device=device)
tokenizer_config = ModelConfig(model_id="google/gemma-3-12b-it-qat-q4_0-unquantized") if tokenizer_path is None else ModelConfig(tokenizer_path)
self.pipe = LTX2AudioVideoPipeline.from_pretrained(torch_dtype=torch.bfloat16, device=device, model_configs=model_configs, tokenizer_config=tokenizer_config)
self.pipe = self.split_pipeline_units(task, self.pipe, trainable_models, lora_base_model)
# Training mode
self.switch_pipe_to_training_mode(
self.pipe, trainable_models,
lora_base_model, lora_target_modules, lora_rank, lora_checkpoint,
preset_lora_path, preset_lora_model,
task=task,
)
# Store other configs
self.use_gradient_checkpointing = use_gradient_checkpointing
self.use_gradient_checkpointing_offload = use_gradient_checkpointing_offload
self.extra_inputs = extra_inputs.split(",") if extra_inputs is not None else []
self.fp8_models = fp8_models
self.task = task
self.task_to_loss = {
"sft:data_process": lambda pipe, *args: args,
"sft": lambda pipe, inputs_shared, inputs_posi, inputs_nega: FlowMatchSFTAudioVideoLoss(pipe, **inputs_shared, **inputs_posi),
"sft:train": lambda pipe, inputs_shared, inputs_posi, inputs_nega: FlowMatchSFTAudioVideoLoss(pipe, **inputs_shared, **inputs_posi),
}
def parse_extra_inputs(self, data, extra_inputs, inputs_shared):
for extra_input in extra_inputs:
inputs_shared[extra_input] = data[extra_input]
return inputs_shared
def get_pipeline_inputs(self, data):
inputs_posi = {"prompt": data["prompt"]}
inputs_nega = {}
inputs_shared = {
# Assume you are using this pipeline for inference,
# please fill in the input parameters.
"input_video": data["video"],
"height": data["video"][0].size[1],
"width": data["video"][0].size[0],
"num_frames": len(data["video"]),
"frame_rate": data.get("frame_rate", 24),
# Please do not modify the following parameters
# unless you clearly know what this will cause.
"cfg_scale": 1,
"tiled": False,
"rand_device": self.pipe.device,
"use_gradient_checkpointing": self.use_gradient_checkpointing,
"use_gradient_checkpointing_offload": self.use_gradient_checkpointing_offload,
"video_patchifier": self.pipe.video_patchifier,
"audio_patchifier": self.pipe.audio_patchifier,
}
inputs_shared = self.parse_extra_inputs(data, self.extra_inputs, inputs_shared)
return inputs_shared, inputs_posi, inputs_nega
def forward(self, data, inputs=None):
if inputs is None: inputs = self.get_pipeline_inputs(data)
inputs = self.transfer_data_to_device(inputs, self.pipe.device, self.pipe.torch_dtype)
for unit in self.pipe.units:
inputs = self.pipe.unit_runner(unit, self.pipe, *inputs)
loss = self.task_to_loss[self.task](self.pipe, *inputs)
return loss
def ltx2_parser():
parser = argparse.ArgumentParser(description="Simple example of a training script.")
parser = add_general_config(parser)
parser = add_video_size_config(parser)
parser.add_argument("--tokenizer_path", type=str, default=None, help="Path to tokenizer.")
parser.add_argument("--frame_rate", type=float, default=24, help="frame rate of the training videos.")
parser.add_argument("--initialize_model_on_cpu", default=False, action="store_true", help="Whether to initialize models on CPU.")
return parser
if __name__ == "__main__":
parser = ltx2_parser()
args = parser.parse_args()
accelerator = accelerate.Accelerator(
gradient_accumulation_steps=args.gradient_accumulation_steps,
kwargs_handlers=[accelerate.DistributedDataParallelKwargs(find_unused_parameters=args.find_unused_parameters)],
)
video_processor = UnifiedDataset.default_video_operator(
base_path=args.dataset_base_path,
max_pixels=args.max_pixels,
height=args.height,
width=args.width,
height_division_factor=32,
width_division_factor=32,
num_frames=args.num_frames,
time_division_factor=8,
time_division_remainder=1,
)
dataset = UnifiedDataset(
base_path=args.dataset_base_path,
metadata_path=args.dataset_metadata_path,
repeat=args.dataset_repeat,
data_file_keys=args.data_file_keys.split(","),
main_data_operator=video_processor,
special_operator_map={
"input_audio": ToAbsolutePath(args.dataset_base_path) >> LoadAudioWithTorchaudio(duration=float(args.num_frames) / float(args.frame_rate)),
"in_context_videos": RouteByType(operator_map=[
(str, video_processor),
(list, SequencialProcess(video_processor)),
]),
}
)
model = LTX2TrainingModule(
model_paths=args.model_paths,
model_id_with_origin_paths=args.model_id_with_origin_paths,
tokenizer_path=args.tokenizer_path,
trainable_models=args.trainable_models,
lora_base_model=args.lora_base_model,
lora_target_modules=args.lora_target_modules,
lora_rank=args.lora_rank,
lora_checkpoint=args.lora_checkpoint,
preset_lora_path=args.preset_lora_path,
preset_lora_model=args.preset_lora_model,
use_gradient_checkpointing=args.use_gradient_checkpointing,
use_gradient_checkpointing_offload=args.use_gradient_checkpointing_offload,
extra_inputs=args.extra_inputs,
fp8_models=args.fp8_models,
offload_models=args.offload_models,
task=args.task,
device="cpu" if args.initialize_model_on_cpu else accelerator.device,
)
model_logger = ModelLogger(
args.output_path,
remove_prefix_in_ckpt=args.remove_prefix_in_ckpt,
)
launcher_map = {
"sft:data_process": launch_data_process_task,
"direct_distill:data_process": launch_data_process_task,
"sft": launch_training_task,
"sft:train": launch_training_task,
"direct_distill": launch_training_task,
"direct_distill:train": launch_training_task,
}
launcher_map[args.task](accelerator, dataset, model, model_logger, args=args)

47
examples/ltx2/model_training/validate_full/LTX-2-T2AV.py Normal file
View File

@@ -0,0 +1,47 @@
import torch
from diffsynth.pipelines.ltx2_audio_video import LTX2AudioVideoPipeline, ModelConfig
from diffsynth.utils.data.media_io_ltx2 import write_video_audio_ltx2
vram_config = {
"offload_dtype": torch.bfloat16,
"offload_device": "cpu",
"onload_dtype": torch.bfloat16,
"onload_device": "cuda",
"preparing_dtype": torch.bfloat16,
"preparing_device": "cuda",
"computation_dtype": torch.bfloat16,
"computation_device": "cuda",
}
pipe = LTX2AudioVideoPipeline.from_pretrained(
torch_dtype=torch.bfloat16,
device="cuda",
model_configs=[
ModelConfig(model_id="google/gemma-3-12b-it-qat-q4_0-unquantized", origin_file_pattern="model-*.safetensors", **vram_config),
ModelConfig(model_id="DiffSynth-Studio/LTX-2-Repackage", origin_file_pattern="text_encoder_post_modules.safetensors", **vram_config),
ModelConfig(model_id="DiffSynth-Studio/LTX-2-Repackage", origin_file_pattern="video_vae_decoder.safetensors", **vram_config),
ModelConfig(model_id="DiffSynth-Studio/LTX-2-Repackage", origin_file_pattern="audio_vae_decoder.safetensors", **vram_config),
ModelConfig(model_id="DiffSynth-Studio/LTX-2-Repackage", origin_file_pattern="audio_vocoder.safetensors", **vram_config),
ModelConfig(path="./models/train/LTX2-T2AV-full/epoch-4.safetensors", **vram_config),
],
tokenizer_config=ModelConfig(model_id="google/gemma-3-12b-it-qat-q4_0-unquantized"),
)
prompt = "A beautiful sunset over the ocean."
negative_prompt = "blurry, out of focus, overexposed, underexposed, low contrast, washed out colors, excessive noise, grainy texture, poor lighting, flickering, motion blur, distorted proportions, unnatural skin tones, deformed facial features, asymmetrical face, missing facial features, extra limbs, disfigured hands, wrong hand count, artifacts around text, inconsistent perspective, camera shake, incorrect depth of field, background too sharp, background clutter, distracting reflections, harsh shadows, inconsistent lighting direction, color banding, cartoonish rendering, 3D CGI look, unrealistic materials, uncanny valley effect, incorrect ethnicity, wrong gender, exaggerated expressions, wrong gaze direction, mismatched lip sync, silent or muted audio, distorted voice, robotic voice, echo, background noise, off-sync audio, incorrect dialogue, added dialogue, repetitive speech, jittery movement, awkward pauses, incorrect timing, unnatural transitions, inconsistent framing, tilted camera, flat lighting, inconsistent tone, cinematic oversaturation, stylized filters, or AI artifacts."
height, width, num_frames = 512, 768, 121
video, audio = pipe(
prompt=prompt,
negative_prompt=negative_prompt,
seed=43,
height=height,
width=width,
num_frames=num_frames,
tiled=True,
cfg_scale=4.0
)
write_video_audio_ltx2(
video=video,
audio=audio,
output_path='ltx2_onestage.mp4',
fps=24,
audio_sample_rate=24000,
)

56
examples/ltx2/model_training/validate_lora/LTX-2-T2AV-IC-LoRA.py Normal file
View File

@@ -0,0 +1,56 @@
import torch
from diffsynth.pipelines.ltx2_audio_video import LTX2AudioVideoPipeline, ModelConfig
from diffsynth.utils.data.media_io_ltx2 import write_video_audio_ltx2
from diffsynth.utils.data import VideoData
vram_config = {
"offload_dtype": torch.bfloat16,
"offload_device": "cpu",
"onload_dtype": torch.bfloat16,
"onload_device": "cuda",
"preparing_dtype": torch.bfloat16,
"preparing_device": "cuda",
"computation_dtype": torch.bfloat16,
"computation_device": "cuda",
}
pipe = LTX2AudioVideoPipeline.from_pretrained(
torch_dtype=torch.bfloat16,
device="cuda",
model_configs=[
ModelConfig(model_id="google/gemma-3-12b-it-qat-q4_0-unquantized", origin_file_pattern="model-*.safetensors", **vram_config),
ModelConfig(model_id="DiffSynth-Studio/LTX-2-Repackage", origin_file_pattern="transformer.safetensors", **vram_config),
ModelConfig(model_id="DiffSynth-Studio/LTX-2-Repackage", origin_file_pattern="text_encoder_post_modules.safetensors", **vram_config),
ModelConfig(model_id="DiffSynth-Studio/LTX-2-Repackage", origin_file_pattern="video_vae_decoder.safetensors", **vram_config),
ModelConfig(model_id="DiffSynth-Studio/LTX-2-Repackage", origin_file_pattern="audio_vae_decoder.safetensors", **vram_config),
ModelConfig(model_id="DiffSynth-Studio/LTX-2-Repackage", origin_file_pattern="audio_vocoder.safetensors", **vram_config),
ModelConfig(model_id="DiffSynth-Studio/LTX-2-Repackage", origin_file_pattern="video_vae_encoder.safetensors", **vram_config),
],
tokenizer_config=ModelConfig(model_id="google/gemma-3-12b-it-qat-q4_0-unquantized"),
)
pipe.load_lora(pipe.dit, "./models/train/LTX2-T2AV-IC-LoRA/epoch-4.safetensors")
prompt = "[VISUAL]:Two cute orange cats, wearing boxing gloves, stand on a boxing ring and fight each other. [SOUNDS]:the sound of two cats boxing"
negative_prompt = "blurry, out of focus, overexposed, underexposed, low contrast, washed out colors, excessive noise, grainy texture, poor lighting, flickering, motion blur, distorted proportions, unnatural skin tones, deformed facial features, asymmetrical face, missing facial features, extra limbs, disfigured hands, wrong hand count, artifacts around text, inconsistent perspective, camera shake, incorrect depth of field, background too sharp, background clutter, distracting reflections, harsh shadows, inconsistent lighting direction, color banding, cartoonish rendering, 3D CGI look, unrealistic materials, uncanny valley effect, incorrect ethnicity, wrong gender, exaggerated expressions, wrong gaze direction, mismatched lip sync, silent or muted audio, distorted voice, robotic voice, echo, background noise, off-sync audio, incorrect dialogue, added dialogue, repetitive speech, jittery movement, awkward pauses, incorrect timing, unnatural transitions, inconsistent framing, tilted camera, flat lighting, inconsistent tone, cinematic oversaturation, stylized filters, or AI artifacts."
height, width, num_frames = 512, 768, 81
ref_scale_factor = 2
frame_rate = 24
input_video = VideoData("data/example_video_dataset/ltx2/depth_video.mp4", height=height // ref_scale_factor // 2, width=width // ref_scale_factor // 2)
input_video = input_video.raw_data()
video, audio = pipe(
prompt=prompt,
negative_prompt=negative_prompt,
seed=43,
height=height,
width=width,
num_frames=num_frames,
frame_rate=frame_rate,
tiled=True,
in_context_videos=[input_video],
in_context_downsample_factor=ref_scale_factor,
)
write_video_audio_ltx2(
video=video,
audio=audio,
output_path='ltx2_onestage_ic.mp4',
fps=frame_rate,
audio_sample_rate=24000,
)

48
examples/ltx2/model_training/validate_lora/LTX-2-T2AV.py Normal file
View File

@@ -0,0 +1,48 @@
import torch
from diffsynth.pipelines.ltx2_audio_video import LTX2AudioVideoPipeline, ModelConfig
from diffsynth.utils.data.media_io_ltx2 import write_video_audio_ltx2
vram_config = {
"offload_dtype": torch.bfloat16,
"offload_device": "cpu",
"onload_dtype": torch.bfloat16,
"onload_device": "cuda",
"preparing_dtype": torch.bfloat16,
"preparing_device": "cuda",
"computation_dtype": torch.bfloat16,
"computation_device": "cuda",
}
pipe = LTX2AudioVideoPipeline.from_pretrained(
torch_dtype=torch.bfloat16,
device="cuda",
model_configs=[
ModelConfig(model_id="google/gemma-3-12b-it-qat-q4_0-unquantized", origin_file_pattern="model-*.safetensors", **vram_config),
ModelConfig(model_id="DiffSynth-Studio/LTX-2-Repackage", origin_file_pattern="transformer.safetensors", **vram_config),
ModelConfig(model_id="DiffSynth-Studio/LTX-2-Repackage", origin_file_pattern="text_encoder_post_modules.safetensors", **vram_config),
ModelConfig(model_id="DiffSynth-Studio/LTX-2-Repackage", origin_file_pattern="video_vae_decoder.safetensors", **vram_config),
ModelConfig(model_id="DiffSynth-Studio/LTX-2-Repackage", origin_file_pattern="audio_vae_decoder.safetensors", **vram_config),
ModelConfig(model_id="DiffSynth-Studio/LTX-2-Repackage", origin_file_pattern="audio_vocoder.safetensors", **vram_config),
],
tokenizer_config=ModelConfig(model_id="google/gemma-3-12b-it-qat-q4_0-unquantized"),
)
pipe.load_lora(pipe.dit, "models/train/LTX2-T2AV_lora/epoch-4.safetensors")
prompt = "A beautiful sunset over the ocean."
negative_prompt = "blurry, out of focus, overexposed, underexposed, low contrast, washed out colors, excessive noise, grainy texture, poor lighting, flickering, motion blur, distorted proportions, unnatural skin tones, deformed facial features, asymmetrical face, missing facial features, extra limbs, disfigured hands, wrong hand count, artifacts around text, inconsistent perspective, camera shake, incorrect depth of field, background too sharp, background clutter, distracting reflections, harsh shadows, inconsistent lighting direction, color banding, cartoonish rendering, 3D CGI look, unrealistic materials, uncanny valley effect, incorrect ethnicity, wrong gender, exaggerated expressions, wrong gaze direction, mismatched lip sync, silent or muted audio, distorted voice, robotic voice, echo, background noise, off-sync audio, incorrect dialogue, added dialogue, repetitive speech, jittery movement, awkward pauses, incorrect timing, unnatural transitions, inconsistent framing, tilted camera, flat lighting, inconsistent tone, cinematic oversaturation, stylized filters, or AI artifacts."
height, width, num_frames = 512, 768, 121
video, audio = pipe(
prompt=prompt,
negative_prompt=negative_prompt,
seed=43,
height=height,
width=width,
num_frames=num_frames,
tiled=True,
cfg_scale=4.0
)
write_video_audio_ltx2(
video=video,
audio=audio,
output_path='ltx2_onestage.mp4',
fps=24,
audio_sample_rate=24000,
)

48
examples/ltx2/model_training/validate_lora/LTX-2-T2AV_noaudio.py Normal file
View File

@@ -0,0 +1,48 @@
import torch
from diffsynth.pipelines.ltx2_audio_video import LTX2AudioVideoPipeline, ModelConfig
from diffsynth.utils.data.media_io_ltx2 import write_video_audio_ltx2
vram_config = {
"offload_dtype": torch.bfloat16,
"offload_device": "cpu",
"onload_dtype": torch.bfloat16,
"onload_device": "cuda",
"preparing_dtype": torch.bfloat16,
"preparing_device": "cuda",
"computation_dtype": torch.bfloat16,
"computation_device": "cuda",
}
pipe = LTX2AudioVideoPipeline.from_pretrained(
torch_dtype=torch.bfloat16,
device="cuda",
model_configs=[
ModelConfig(model_id="google/gemma-3-12b-it-qat-q4_0-unquantized", origin_file_pattern="model-*.safetensors", **vram_config),
ModelConfig(model_id="DiffSynth-Studio/LTX-2-Repackage", origin_file_pattern="transformer.safetensors", **vram_config),
ModelConfig(model_id="DiffSynth-Studio/LTX-2-Repackage", origin_file_pattern="text_encoder_post_modules.safetensors", **vram_config),
ModelConfig(model_id="DiffSynth-Studio/LTX-2-Repackage", origin_file_pattern="video_vae_decoder.safetensors", **vram_config),
ModelConfig(model_id="DiffSynth-Studio/LTX-2-Repackage", origin_file_pattern="audio_vae_decoder.safetensors", **vram_config),
ModelConfig(model_id="DiffSynth-Studio/LTX-2-Repackage", origin_file_pattern="audio_vocoder.safetensors", **vram_config),
],
tokenizer_config=ModelConfig(model_id="google/gemma-3-12b-it-qat-q4_0-unquantized"),
)
pipe.load_lora(pipe.dit, "models/train/LTX2-T2AV-noaudio_lora/epoch-4.safetensors")
prompt = "A beautiful sunset over the ocean."
negative_prompt = "blurry, out of focus, overexposed, underexposed, low contrast, washed out colors, excessive noise, grainy texture, poor lighting, flickering, motion blur, distorted proportions, unnatural skin tones, deformed facial features, asymmetrical face, missing facial features, extra limbs, disfigured hands, wrong hand count, artifacts around text, inconsistent perspective, camera shake, incorrect depth of field, background too sharp, background clutter, distracting reflections, harsh shadows, inconsistent lighting direction, color banding, cartoonish rendering, 3D CGI look, unrealistic materials, uncanny valley effect, incorrect ethnicity, wrong gender, exaggerated expressions, wrong gaze direction, mismatched lip sync, silent or muted audio, distorted voice, robotic voice, echo, background noise, off-sync audio, incorrect dialogue, added dialogue, repetitive speech, jittery movement, awkward pauses, incorrect timing, unnatural transitions, inconsistent framing, tilted camera, flat lighting, inconsistent tone, cinematic oversaturation, stylized filters, or AI artifacts."
height, width, num_frames = 512, 768, 121
video, audio = pipe(
prompt=prompt,
negative_prompt=negative_prompt,
seed=43,
height=height,
width=width,
num_frames=num_frames,
tiled=True,
cfg_scale=4.0
)
write_video_audio_ltx2(
video=video,
audio=audio,
output_path='ltx2_onestage.mp4',
fps=24,
audio_sample_rate=24000,
)

2
examples/qwen_image/model_inference/Qwen-Image-Edit-2511-Lightning.py
View File

@@ -18,7 +18,7 @@ lora = ModelConfig(
model_id="lightx2v/Qwen-Image-Edit-2511-Lightning",
origin_file_pattern="Qwen-Image-Edit-2511-Lightning-4steps-V1.0-bf16.safetensors"
)
pipe.load_lora(pipe.dit, lora, alpha=8/64)
pipe.load_lora(pipe.dit, lora, alpha=1)
pipe.scheduler = FlowMatchScheduler("Qwen-Image-Lightning")

2
examples/qwen_image/model_inference_low_vram/Qwen-Image-Edit-2511-Lightning.py
View File

@@ -28,7 +28,7 @@ lora = ModelConfig(
model_id="lightx2v/Qwen-Image-Edit-2511-Lightning",
origin_file_pattern="Qwen-Image-Edit-2511-Lightning-4steps-V1.0-bf16.safetensors"
)
pipe.load_lora(pipe.dit, lora, alpha=8/64)
pipe.load_lora(pipe.dit, lora, alpha=1)
pipe.scheduler = FlowMatchScheduler("Qwen-Image-Lightning")