Instructions to use akshan-main/tiny-ltx-modular-pipe with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- Diffusers
How to use akshan-main/tiny-ltx-modular-pipe with Diffusers:
pip install -U diffusers transformers accelerate
import torch from diffusers import DiffusionPipeline # switch to "mps" for apple devices pipe = DiffusionPipeline.from_pretrained("akshan-main/tiny-ltx-modular-pipe", dtype=torch.bfloat16, device_map="cuda") prompt = "Astronaut in a jungle, cold color palette, muted colors, detailed, 8k" image = pipe(prompt).images[0] - LTX.io
How to use akshan-main/tiny-ltx-modular-pipe with LTX.io:
# Install the LTX-2 pipelines git clone https://github.com/Lightricks/LTX-2.git cd LTX-2 uv sync --frozen
# Download the weights from this repo, plus the Gemma text encoder hf download akshan-main/tiny-ltx-modular-pipe --local-dir models/tiny-ltx-modular-pipe hf download google/gemma-3-12b-it-qat-q4_0-unquantized --local-dir models/gemma-3-12b
# Fast pipeline (distilled model, no distilled LoRA needed) uv run python -m ltx_pipelines.distilled \ --distilled-checkpoint-path models/tiny-ltx-modular-pipe/<distilled-checkpoint>.safetensors \ --spatial-upsampler-path models/tiny-ltx-modular-pipe/<spatial-upsampler>.safetensors \ --gemma-root models/gemma-3-12b \ --prompt "A beautiful sunset over the ocean" \ --output-path output.mp4 # For image-to-video, add: --image path/to/image.jpg 0 0.8# HQ pipeline (two-stage, higher quality) uv run python -m ltx_pipelines.ti2vid_two_stages_hq \ --checkpoint-path models/tiny-ltx-modular-pipe/<checkpoint>.safetensors \ --distilled-lora models/tiny-ltx-modular-pipe/<distilled-lora>.safetensors 0.8 \ --spatial-upsampler-path models/tiny-ltx-modular-pipe/<spatial-upsampler>.safetensors \ --gemma-root models/gemma-3-12b \ --prompt "A beautiful sunset over the ocean" \ --output-path output.mp4 # For image-to-video, add: --image path/to/image.jpg 0 0.8 - Notebooks
- Google Colab
- Kaggle
| { | |
| "architectures": [ | |
| "T5EncoderModel" | |
| ], | |
| "classifier_dropout": 0.0, | |
| "d_ff": 64, | |
| "d_kv": 8, | |
| "d_model": 32, | |
| "decoder_start_token_id": 0, | |
| "dense_act_fn": "relu", | |
| "dropout_rate": 0.1, | |
| "dtype": "float32", | |
| "eos_token_id": 1, | |
| "feed_forward_proj": "relu", | |
| "initializer_factor": 1.0, | |
| "is_encoder_decoder": false, | |
| "is_gated_act": false, | |
| "layer_norm_epsilon": 1e-06, | |
| "model_type": "t5", | |
| "num_decoder_layers": 1, | |
| "num_heads": 4, | |
| "num_layers": 1, | |
| "pad_token_id": 0, | |
| "relative_attention_max_distance": 128, | |
| "relative_attention_num_buckets": 32, | |
| "transformers_version": "4.57.3", | |
| "use_cache": false, | |
| "vocab_size": 32100 | |
| } | |