| title: LTX-2.3 Video Generator | |
| emoji: 🎬 | |
| colorFrom: purple | |
| colorTo: pink | |
| sdk: gradio | |
| sdk_version: 5.25.0 | |
| python_version: "3.12" | |
| app_file: app.py | |
| pinned: false | |
| short_description: Text-to-Video & Image-to-Video with Lightricks LTX-2.3 | |
| # 🎬 LTX-2.3 Video Generator | |
| Generate videos from text prompts or images using **Lightricks LTX-2.3** — the latest open-source audio-video foundation model. | |
| **Features:** | |
| - ✏️ **Text-to-Video** — describe any scene | |
| - 🖼️ **Image-to-Video** — animate a starting image | |
| - 🎞️ Adjustable frames, steps, guidance, resolution | |
| - 🎲 Seed control for reproducible results | |
| - ⚡ Uses the distilled 22B checkpoint for fast generation | |
| ### Hardware Requirements | |
| - Runs on **ZeroGPU** (48GB Blackwell) or any NVIDIA GPU with 24GB+ VRAM | |
| - Uses FP8 / bfloat16 for memory efficiency | |
| ### Model | |
| [Lightricks/LTX-2.3](https://huggingface.co/Lightricks/LTX-2.3) — 22B parameter DiT model | |