text-to-video and image-to-video generation (diffusion models)
Official code for Lightricks' LTX family of text-to-video and image-to-video diffusion models, including the latest LTX-2 with synchronized audio+video generation.
Solves the same problem with a different architecture or at a different layer. Expect to rewrite the integration.
choose when you need a dedicated image-to-video model derived from HunyuanVideo, especially in a Tencent ecosystem workflow.
HunyuanVideo-I2V: A Customizable Image-to-Video Model based on HunyuanVideo
choose if you need a lightweight video-generation inference framework emphasizing speed/efficiency for resource-constrained deployment.
Lightweight Image Video Action Generation Inference Framework
choose for a lightweight Tencent video-generation model if you prefer its ecosystem, though it lacks the audio-sync capabilities of LTX-2.
HunyuanVideo-1.5: A leading lightweight video generation model
choose if you prefer a different open-source foundation model that unifies audio+video generation and want to avoid LTX's codebase.
A foundation model that generates synchronized video and audio in a single model
choose when you need real-time long-video generation from PKU, if its model weights fit your needs.
Helios: Real Real-Time Long Video Generation Model
choose for a diffusion family covering both image and video if you prefer Kandinsky's trained models and the MIT license.
Kandinsky 5.0: A family of diffusion models for Video & Image generation
An earlier or more limited way to do the job. Still the right call when you need something small, proven, or CPU-only.
narrower: focuses on customized subject-driven video generation rather than general T2V/I2V.
HunyuanCustom: A Multimodal-Driven Architecture for Customized Video Generation
older research baseline (ICLR 2024) with LLM-grounded prompting; use if you need reproduction of that method rather than a maintained model.
[ICLR 2024] LLM-grounded Video Diffusion Models (LVD): official implementation for the LVD paper
narrower: choose when animating a single still image with explicit text-guided motion control is the primary need.
Official implementations for paper: LivePhoto: Real Image Animation with Text-guided Motion Control
narrower: portrait animation with implicit condition control; use if you only need portrait-specific animation.
[CVPR-2025] The official code of HunyuanPortrait: Implicit Condition Control for Enhanced Portrait Animation
narrower: video generation focused on subject consistency; useful for object-centric consistency rather than open text-to-video.
Phantom: Subject-Consistent Video Generation via Cross-Modal Alignment
Does the same job, but ships as an app. Useful to a person, not swappable into a codebase.
select when you want a local GUI studio for AI video/music creation; not library-swappable.
An all-in-one, 100% local AI video, image, and music studio. Director mode plans full music videos and short films from a single prompt. Built on the WanGP pipeline. Install via Pinokio.
These projects were analysed and named LTX-Video among their alternatives. The relationship is not symmetric — how LTX-Video rates them is a separate judgement, made when LTX-Video is analysed in its own right.
calls LTX-Video “Pick it when you need fast, low-VRAM generation under a permissive Apache-2.0 license.”
text and image to video generation: CogVideoX (2024) and CogVideo (ICLR 2023)
calls LTX-Video “use it when inference speed matters, as it targets real-time generation under Apache-2.0”
HunyuanVideo: A Systematic Framework For Large Video Generation Model
calls LTX-Video “Choose when you need LTX-Video's real-time text-to-video model family and weights instead of Sana-Video.”
SANA: Efficient High-Resolution Image Synthesis with Linear Diffusion Transformer
calls LTX-Video “Choose when you need the specific LTX-Video architecture for text-to-video generation.”
OpenMMLab Multimodal Advanced, Generative, and Intelligent Creation Toolbox. Unlock the magic 🪄: Generative-AI (AIGC), easy-to-use APIs, awsome model zoo, diffusion models, for text-to-image generation, image/video restoration/enhancement, etc.