Instructions to use kabachuha/ltx2-eat with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- Diffusers
How to use kabachuha/ltx2-eat with Diffusers:
pip install -U diffusers transformers accelerate
import torch from diffusers import DiffusionPipeline from diffusers.utils import load_image, export_to_video # switch to "mps" for apple devices pipe = DiffusionPipeline.from_pretrained("Lightricks/LTX-2", dtype=torch.bfloat16, device_map="cuda") pipe.load_lora_weights("kabachuha/ltx2-eat") prompt = "Vivid colors. The video begins with a cartoon girl blowing a chewing gum balloon. Then a gigantic anime girl hand seizes the cartoon girl from below and tosses her into a gigantic mouth, which appeared to the right. The camera zooms out, showing the new anime girl chewing and fully swallowing the cartoon girl. Wet slurping sounds." input_image = load_image("https://huggingface.co/datasets/huggingface/documentation-images/resolve/main/diffusers/guitar-man.png") image = pipe(image=input_image, prompt=prompt).frames[0] export_to_video(output, "output.mp4") - Inference
- Notebooks
- Google Colab
- Kaggle
- Local Apps Settings
- Draw Things
| { | |
| "caption_dropout_probability": 0.0, | |
| "checkpoint_epoch_interval": 4, | |
| "checkpoints_total_limit": 99, | |
| "compress_disk_cache": true, | |
| "data_backend_config": "/media/kabachuha/xiangliu/simpletunerx/ltx2-multiresolution-eat-t2v-v2.json", | |
| "disable_benchmark": true, | |
| "hub_model_id": "ltx2-lora-eat", | |
| "push_checkpoints_to_hub": false, | |
| "push_to_hub": false, | |
| "peft_lora_target_modules": ["to_k", "to_q", "to_v", "to_out.0", "ff.net.0.proj", "ff.net.2"], | |
| "quantize_activations": false, | |
| "gradient_checkpointing": true, | |
| "musubi_blocks_to_swap": 0, | |
| "learning_rate": 1.0, | |
| "lora_rank": 64, | |
| "lora_type": "standard", | |
| "lora_format": "comfyui", | |
| "lr_scheduler": "constant", | |
| "lr_warmup_steps": 10, | |
| "max_grad_norm": 1.0, | |
| "num_train_epochs": 120, | |
| "prodigy_steps": 100000, | |
| "minimum_image_size": 0, | |
| "validation_disable": true, | |
| "offload_during_save": false, | |
| "base_model_precision": "int8-quanto", | |
| "model_family": "ltxvideo2", | |
| "quantize_via": "cpu", | |
| "model_flavour": "dev", | |
| "pretrained_model_name_or_path": "/media/kabachuha/xiangliu/LTX2/ltx-2-19b-dev.safetensors", | |
| "pretrained_gemma_model_name_or_path": "/media/kabachuha/xiangliu/LTX2/gemma-3-12b-it-qat-q4_0-unquantized", | |
| "ramtorch": false, | |
| "ramtorch_text_encoder": false, | |
| "model_type": "lora", | |
| "optimizer": "prodigy", | |
| "loss_type":"huber", | |
| "huber_c": 1.0, | |
| "huber_schedule": "exponential", | |
| "output_dir": "/media/kabachuha/xiangliu/simpletunerx/ltx2-models-eat-v2-sphl-v2", | |
| "report_to": "tensorboard", | |
| "resolution": 480, | |
| "resolution_type": "pixel_area", | |
| "seed": 6934, | |
| "vae_enable_patch_conv": true, | |
| "tracker_project_name": "ltx2-lora-eat", | |
| "tracker_run_name": "ltx2-training-eat", | |
| "train_batch_size": 1, | |
| "vae_batch_size": 1, | |
| "vae_enable_tiling": true, | |
| "vae_enable_slicing": true, | |
| "vae_enable_temporal_roll": true, | |
| "validation_guidance": 5.0, | |
| "validation_negative_prompt": "blurry, cropped, ugly", | |
| "validation_num_inference_steps": 20, | |
| "validation_num_video_frames": 73, | |
| "validation_prompt": "eat style. The video starts with a close-up of a cute anime girl. Suddenly, a colossal anime girl's hand swoops down and scoops her up. The camera pans up to reveal the colossal girl's gaping maw, filled with rows of sharp teeth. The tiny anime girl is tossed directly into the mouth and disappears instantly, swallowed whole. The colossal girl lets out a satisfied moan, her breasts jiggling as she chews.", | |
| "validation_prompt_library": false, | |
| "validation_resolution": "480x640", | |
| "validation_seed": 6934, | |
| "validation_steps": 50, | |
| "offload_during_startup": true, | |
| "flow_schedule_shift": 2.0, | |
| "resume_from_checkpoint": "latest", | |
| "crepa_enabled": false, | |
| "crepa_block_index": 8, | |
| "crepa_lambda": 0.02, | |
| "crepa_adjacent_distance": 1, | |
| "crepa_adjacent_tau": 1.0, | |
| "crepa_use_backbone_features": true, | |
| "crepa_teacher_block_index": 42, | |
| "crepa_drop_vae_encoder": false, | |
| "crepa_encoder": "dinov2_vitb14", | |
| "crepa_encoder_image_size": 544, | |
| "tread_config": { | |
| "routes": [ | |
| { | |
| "selection_ratio": 0.25, | |
| "start_layer_idx": 2, | |
| "end_layer_idx": -2 | |
| } | |
| ] | |
| } | |
| } | |