LTX 2 Distilled
Neural Network
LTX 2 Distilled is an open audio-video model for generating video with sound from text descriptions.
Specifications
- Input
- text, images
- Output
- video
- Aspect ratio
- 16:9, 9:16, 4:3, 3:4, 1:1, 21:9
Video generation
(duration 8 sec., 720p)
Video generation
(duration 8 sec., 1080p)
How it works LTX 2 Distilled?
LTX-2 is an open audio-video model: sound appears alongside the image rather than being added as a separate step, so the video comes with a ready-made soundtrack. Distilled is a lightweight version of LTX-2, and the open code provides a clear foundation for product scenarios. It works simply: you describe a scene, and the model generates the frame and sound. In BotHub, the model is accessible from Russia without a VPN or foreign card; payment is via Russian cards in rubles, only for actual generations, with no subscriptions or expiring packages. With over 250 neural networks in one window, you can easily run the same scenario across multiple models to compare. For integrations, there is a unified OpenAI-compatible API; for companies, we offer contracts, invoices, electronic document management, and an admin panel with limits. It is useful when you need a voiced draft of a commercial before filming, when social media requires new videos weekly, when you need dynamic inserts and intros for presentations or online courses, and for storyboards and scene tests before major production.Frequently asked questions about LTX 2 Distilled
You can use generated results for commercial purposes. You own all rights to the content you create. The only restriction: make sure your prompt does not include copyrighted third-party material. You are responsible for respecting the rights to any input data.
ltx-2-distilled by Lightricks is a video model: it generates a clip from a text description or animates an uploaded image. Suitable for short social media clips, ad intros, concept animation, and scene pre-visualization. Accessible from Russia without a VPN or foreign card, payment in rubles.
You get a finished video clip that can be downloaded directly from the BotHub interface or retrieved via API. See the technical header above for exact duration and resolution parameters — we won't speculate on them for the model.
A separate step-by-step reasoning mode is not noted in our data — this is a generative video model, not a chat assistant. Result quality depends more on the prompt: describe the scene, action, camera movement, lighting, and mood, and the frame will be closer to your vision.
Function calling and structured JSON output are not noted in our data: the model returns video, not a text object. It is convenient to keep orchestration logic on a text model — BotHub's unified OpenAI-compatible API allows switching between them without rewriting the integration.
Image — yes: the image-to-video mode animates an uploaded frame while preserving its composition, character, and style, which is convenient for animating ready-made layouts. Inputting audio or video is not noted in our data, so the main scenario is text plus a reference image.
We do not track language-specific prompt features for this model in our data, so we won't make any promises: test it with your task. BotHub itself is Russian-speaking — the interface, support, and payment with Russian cards work without a VPN, and tokens do not expire and are charged based on actual usage.