Instructions to use oxide-lab/LTX-Video-0.9.5-diffusers with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use oxide-lab/LTX-Video-0.9.5-diffusers with Diffusers:

pip install -U diffusers transformers accelerate

import torch
from diffusers import DiffusionPipeline

# switch to "mps" for apple devices
pipe = DiffusionPipeline.from_pretrained("oxide-lab/LTX-Video-0.9.5-diffusers", dtype=torch.bfloat16, device_map="cuda")

prompt = "Astronaut in a jungle, cold color palette, muted colors, detailed, 8k"
image = pipe(prompt).images[0]

llama-cpp-python

How to use oxide-lab/LTX-Video-0.9.5-diffusers with llama-cpp-python:

# !pip install llama-cpp-python

from llama_cpp import Llama

llm = Llama.from_pretrained(
	repo_id="oxide-lab/LTX-Video-0.9.5-diffusers",
	filename="text_encoder_gguf/t5-v1_1-xxl-encoder-Q5_K_M.gguf",
)

output = llm(
	"Once upon a time,",
	max_tokens=512,
	echo=True
)
print(output)

Notebooks
Google Colab
Kaggle
Local Apps

llama.cpp

How to use oxide-lab/LTX-Video-0.9.5-diffusers with llama.cpp:

Install from brew

brew install llama.cpp
# Start a local OpenAI-compatible server with a web UI:
llama-server -hf oxide-lab/LTX-Video-0.9.5-diffusers:Q5_K_M
# Run inference directly in the terminal:
llama-cli -hf oxide-lab/LTX-Video-0.9.5-diffusers:Q5_K_M

Install from WinGet (Windows)

winget install llama.cpp
# Start a local OpenAI-compatible server with a web UI:
llama-server -hf oxide-lab/LTX-Video-0.9.5-diffusers:Q5_K_M
# Run inference directly in the terminal:
llama-cli -hf oxide-lab/LTX-Video-0.9.5-diffusers:Q5_K_M

Use pre-built binary

# Download pre-built binary from:
# https://github.com/ggerganov/llama.cpp/releases
# Start a local OpenAI-compatible server with a web UI:
./llama-server -hf oxide-lab/LTX-Video-0.9.5-diffusers:Q5_K_M
# Run inference directly in the terminal:
./llama-cli -hf oxide-lab/LTX-Video-0.9.5-diffusers:Q5_K_M

Build from source code

git clone https://github.com/ggerganov/llama.cpp.git
cd llama.cpp
cmake -B build
cmake --build build -j --target llama-server llama-cli
# Start a local OpenAI-compatible server with a web UI:
./build/bin/llama-server -hf oxide-lab/LTX-Video-0.9.5-diffusers:Q5_K_M
# Run inference directly in the terminal:
./build/bin/llama-cli -hf oxide-lab/LTX-Video-0.9.5-diffusers:Q5_K_M

Use Docker

docker model run hf.co/oxide-lab/LTX-Video-0.9.5-diffusers:Q5_K_M

LM Studio
Jan
Ollama
How to use oxide-lab/LTX-Video-0.9.5-diffusers with Ollama:
```
ollama run hf.co/oxide-lab/LTX-Video-0.9.5-diffusers:Q5_K_M
```

Unsloth Studio new

How to use oxide-lab/LTX-Video-0.9.5-diffusers with Unsloth Studio:

Install Unsloth Studio (macOS, Linux, WSL)

curl -fsSL https://unsloth.ai/install.sh | sh
# Run unsloth studio
unsloth studio -H 0.0.0.0 -p 8888
# Then open http://localhost:8888 in your browser
# Search for oxide-lab/LTX-Video-0.9.5-diffusers to start chatting

Install Unsloth Studio (Windows)

irm https://unsloth.ai/install.ps1 | iex
# Run unsloth studio
unsloth studio -H 0.0.0.0 -p 8888
# Then open http://localhost:8888 in your browser
# Search for oxide-lab/LTX-Video-0.9.5-diffusers to start chatting

Using HuggingFace Spaces for Unsloth

# No setup required
# Open https://huggingface.co/spaces/unsloth/studio in your browser
# Search for oxide-lab/LTX-Video-0.9.5-diffusers to start chatting

Docker Model Runner
How to use oxide-lab/LTX-Video-0.9.5-diffusers with Docker Model Runner:
```
docker model run hf.co/oxide-lab/LTX-Video-0.9.5-diffusers:Q5_K_M
```

Lemonade

How to use oxide-lab/LTX-Video-0.9.5-diffusers with Lemonade:

Pull the model

# Download Lemonade from https://lemonade-server.ai/
lemonade pull oxide-lab/LTX-Video-0.9.5-diffusers:Q5_K_M

Run and chat with the model

lemonade run user.LTX-Video-0.9.5-diffusers-Q5_K_M

List all available models

lemonade list

LTX-Video in Rust (Candle)

This repository provides a high-performance, native Rust implementation of LTX-Video using the Candle ML framework.

Features

🦀 Native Rust: No Python dependency required for inference.
🚀 Performance: Optimized for NVIDIA GPUs with Flash Attention v2 and cuDNN.
💾 Memory Efficient: Supports GGUF quantization for T5-XXL text encoder and VAE tiling/slicing for generating HD videos on consumer GPUs.
🛠 Flexible: Easy to use CLI for video generation and library for custom integration.

Quick Start

Installation

Ensure you have Rust and the CUDA Toolkit installed, then:

git clone https://github.com/FerrisMind/candle-video
cd candle-video
cargo build --release --features flash-attn,cudnn

Video Generation

cargo run --example ltx-video --release -- \
    --local-weights ./models/ltx-video \
    --prompt "A serene mountain lake at sunset, photorealistic, 4k" \
    --width 768 --height 512 --num-frames 97 \
    --steps 30

Performance & Memory

Resolution	Frames	VRAM (BF16)	VRAM (VAE Tiling)
512x768	97	~8-13 GB	~8-9 GB

Note: Using GGUF T5 encoder saves an additional ~8-12GB of VRAM.

Credits

Original Model: Lightricks/LTX-Video
Framework: HuggingFace Candle
T5 v1_1 XXl GGUF and Safetensors: city96/LTX-Video-gguf (for GGUF support patterns, T5 XXl GGUF and Safetensors)

For more details, visit the main GitHub Repository.

Downloads last month: 209

GGUF

Model size

5B params

Architecture

t5encoder

Hardware compatibility

5-bit

Model tree for oxide-lab/LTX-Video-0.9.5-diffusers

Base model

Lightricks/LTX-Video-0.9.5

Quantized

(1)

this model

Collection including oxide-lab/LTX-Video-0.9.5-diffusers

LTX-Video

Collection

LTX-Video 0.9.5+ model weights for candle-video • 2 items • Updated Jan 9