fal-ai/ltx-video/image-to-video

Generate videos from images using LTX Video

Inference

Research only

Input

Prompt*

A lone astronaut in a white spacesuit with gold-tinted visor drifts weightlessly through a sleek, cylindrical corridor of a spaceship. Their movements are slow and graceful as they gently push off the metallic walls with their gloved hands, rotating slightly as they float from right to left across the frame. The corridor features brushed aluminum panels with blue LED strips running along the ceiling, casting a cool glow on the astronaut's suit. Various cables, pipes, and control panels line the walls. The camera follows the astronaut's movement in a handheld style, slightly swaying and adjusting focus, maintaining a medium shot that captures both the astronaut and the corridor's depth. Small particles of dust catch the light as they float in the zero-gravity environment. The scene appears cinematic, with lens flares occasionally reflecting off the metallic surfaces and the astronaut's visor.

Type # to reference inputs.

Image URL*

Hint: Drag and drop image files from your computer, images from web pages, paste from clipboard (Ctrl/Cmd+V), or provide a URL. Accepted file types: jpg, jpeg, png, webp, gif, avif, heic, heif

Additional Settings

Customize your input with more control.

Result

Idle

What would you like to do next?

Your request will cost $0.02 per video.

LTX Video | [image-to-video]

Lightricks' LTX Video transforms static images into 5-second video clips at $0.02 per generation. Trading extended duration for rapid inference and cost efficiency, this model processes image-to-video requests with 30 inference steps and configurable guidance scaling. Built for creators needing quick video extensions from existing imagery without frame-by-frame animation workflows.

Use Cases: Social Media Content Creation | Product Demo Animation | Storyboard Prototyping

Performance

At $0.02 per video generation, LTX Video delivers 50 generations per dollar, positioning it as a cost-efficient entry point for image-to-video workflows compared to premium alternatives ranging from $0.05-$0.15+ per inference.

Metric	Result	Context
Output Duration	5 seconds	Standard for rapid social content iteration
Inference Steps	30 (configurable 1-50)	Balances quality vs speed; adjustable via API
Cost per Video	$0.02	50 generations per $1.00 on fal
Guidance Scale	2-10 (default 3)	Controls prompt adherence strength
Related Endpoints	Text-to-Video	Text-based generation without source image

Image-Driven Motion Synthesis at API Scale

LTX Video applies latent diffusion to the temporal dimension, extending single frames into short-form video through learned motion patterns rather than explicit keyframe animation. Unlike text-only video models that generate from scratch, this architecture preserves source image composition while synthesizing camera movement, object motion, and environmental dynamics.

What this means for you:

Source Image Fidelity: Maintains visual consistency with uploaded imagery, character appearance, lighting conditions, and scene composition carry through to generated motion
Prompt-Guided Motion Control: Natural language descriptions direct movement patterns (e.g., "drifts weightlessly," "slow and graceful") without technical animation parameters
Negative Prompt Refinement: Built-in quality filters suppress common artifacts (motion smear, anatomy distortions, fused fingers) through inverse guidance
Deterministic Generation: Seed parameter enables reproducible outputs for iterative refinement workflows

Technical Specifications

Spec	Details
Architecture	LTX Video
Input Formats	Image URL (JPEG, PNG, WebP, GIF, AVIF) + text prompt
Output Formats	Video file (5-second duration)
Inference Control	1-50 steps, guidance scale 2-10
License	Research only

API Documentation | Quickstart Guide | Enterprise Pricing

How It Stacks Up

Kling Video v2.6 Image to Video ($0.05) – LTX Video trades extended duration and resolution options for 2.5x cost efficiency at $0.02 per generation. Kling v2.6 prioritizes longer-form content (up to 10 seconds) and higher fidelity for polished commercial work where production value justifies premium pricing.

Pixverse Image to Video ($0.05) – LTX Video offers faster iteration cycles through lower per-generation cost, ideal for rapid prototyping and social content workflows. Pixverse emphasizes stylistic control and visual effects integration for creators requiring advanced motion customization at $0.05 per inference.

LongCat Video Image to Video ($0.03) – LTX Video delivers comparable pricing with streamlined parameter configuration (fewer adjustment variables). LongCat provides 720p output specification and extended control surfaces for users needing explicit resolution targeting at slightly higher cost.

MiniMax Hailuo 2.3 Pro ($0.15) – LTX Video prioritizes cost efficiency for volume workflows at 7.5x lower pricing. Hailuo 2.3 Pro targets premium production requirements with advanced temporal coherence and extended duration capabilities where per-generation cost is secondary to output quality.