Fast previsualization
Generate short cinematic concepts quickly before committing time or credits to a higher-fidelity workflow.
Independent tool overview
LTX-2.5 Fast (successor to LTX-2-Fast) at a glance
The original LTX-2-Fast API model was retired in August 2026. Its current successor is LTX-2.5 Fast, Lightricks' speed-focused audio-video model for text-to-video, image-to-video, and audio-to-video generation with synchronized sound, automatic duration, portrait or landscape output, and resolutions up to 4K.
Visit the official LTX-2.5 Fast (successor to LTX-2-Fast) site ↗
- Current model ID
- ltx-2-5-fast
- Original model ID
- ltx-2-fast — retired
- Inputs
- Text, image, or audio
- Maximum output
- 4K at up to 50 fps
- Clip duration
- Up to 20 seconds at 720p or 1080p
- API price
- $0.09–$0.30 per output second
Overview
What LTX-2.5 Fast (successor to LTX-2-Fast) is
LTX-2.5 Fast is the practical migration target for creators and developers who previously used the ltx-2-fast model ID. The old ID is no longer accepted; current API requests must use ltx-2-5-fast or another supported LTX-2.3 or LTX-2.5 model.
The Fast variant prioritizes speed, lower per-second cost, and a wider resolution range than LTX-2.5 Pro. It supports text, image, or audio conditioning, generates synchronized audio with the video, and can produce connected multi-shot scenes while maintaining more consistency across cuts.
There are three distinct access paths: the LTX API for developers, LTX Studio for browser-based creative work, and downloadable model code and weights for technically capable local or cloud environments. Pricing and licensing differ between those paths.
Use cases
Who LTX-2.5 Fast (successor to LTX-2-Fast) is best for
The strongest fit depends on the job you need the product to complete, not the size of its feature list.
Synchronized audio-video
Create visuals and matching sound together instead of building separate silent-video and audio pipelines.
Automated video generation
Use sync or async API endpoints for text-to-video, image-to-video, and audio-driven video in applications and content pipelines.
High-resolution delivery
Generate portrait or landscape output at 720p, 1080p, 1440p, or 4K when the Fast variant's quality is sufficient.
Capabilities
Core LTX-2.5 Fast (successor to LTX-2-Fast) features
Text-to-video
Generate a video and matching audio from a prompt describing scene, movement, camera, dialogue, and mood.
Image-to-video
Animate an input image, optionally constrain the ending with a last frame, and specify camera movement.
Audio-to-video
Generate visuals synchronized to an audio input, with up to 20 seconds at 720p or 1080p and 10 seconds at higher resolutions.
Automatic duration
Set duration to null and let LTX-2.5 choose an available clip length from the prompt, with billing based on the actual output.
Native multi-shot scenes
Create multiple connected shots in one generation while aiming to preserve character, setting, lighting, style, and voice.
Up to 4K and 50 fps
Fast supports portrait and landscape video through 4K, with available frame rates and duration limits varying by resolution.
Open-weight workflow
Use Lightricks' public inference repository, current gated checkpoints, ComfyUI integration, and LoRA training tools in a self-managed environment.
Process
How the LTX-2.5 Fast (successor to LTX-2-Fast) workflow works
Step 1
Choose the access path
Use LTX Studio for a visual workspace, the API for application integration, or the public model repository for self-managed inference.
Step 2
Migrate the model ID
Replace ltx-2-fast with ltx-2-5-fast and verify endpoint, resolution, duration, frame-rate, and billing assumptions.
Step 3
Design a shot-aware prompt
Describe subject, action, environment, camera movement, lighting, sound, dialogue, and transitions in chronological order.
Step 4
Pick resolution and duration
Start at 720p or 1080p for iteration, use a fixed duration when timing matters, and reserve higher resolution for approved shots.
Step 5
Generate and review
Check motion, anatomy, continuity, lip synchronization, text, audio artifacts, safety, and brand or rights issues frame by frame.
Step 6
Control production cost
Calculate per-second API cost before batch generation, monitor Studio credits separately, and keep retries and discarded variations in the budget.
Cost
LTX-2.5 Fast (successor to LTX-2-Fast) pricing and free plan
The LTX API bills LTX-2.5 Fast per output second: $0.09 at 720p, $0.13 at 1080p, $0.19 at 1440p, and $0.30 at 4K. LTX Studio is separate: Free includes 800 one-time credits, Lite is $15 monthly, Standard is $35, and Pro is $125. Annual billing is $12, $28, and $100 per month respectively.
LTX-2.5 Fast API — 720p
$0.09/second
Text-, image-, or audio-to-video at 1280×720 or 720×1280.
- 6–20 seconds at 24/25 fps
- 6–10 seconds at 48/50 fps
- Billed by output duration for text and image generation
LTX-2.5 Fast API — 1080p
$0.13/second
Full-HD portrait or landscape output.
- 6–20 seconds at 24/25 fps
- 6–10 seconds at 48/50 fps
- Audio-to-video billed by input-audio duration
LTX-2.5 Fast API — 1440p
$0.19/second
Higher-resolution output for short approved shots.
- 6, 8, or 10 seconds
- 24, 25, 48, or 50 fps
- Portrait and landscape
LTX-2.5 Fast API — 4K
$0.30/second
Maximum listed Fast resolution.
- 6, 8, or 10 seconds
- 24, 25, 48, or 50 fps
- A 10-second output costs $3 before retries
LTX Studio
$0–$125/month
Browser workspace with model access, storyboarding, editing, and credit allowances.
- Free: 800 one-time credits, personal use
- Lite: $15 monthly and 8,000 credits, personal use
- Standard: $35 and 28,000 credits, commercial use
- Pro: $125 and 110,000 credits, commercial use
- Annual plans receive listed 20% discount
Self-hosted weights
No subscription
Run public checkpoints and inference code under their license terms.
- User supplies GPU compute and storage
- Gated model terms apply
- Engineering and infrastructure costs vary
Pricing checked . Check current pricing at the source ↗
Assessment
LTX-2.5 Fast (successor to LTX-2-Fast) strengths and limitations
Where it stands out
- One model generates synchronized video and audio
- Fast variant supports text, image, and audio inputs
- Portrait and landscape output from 720p through 4K
- Automatic duration and explicit camera-motion controls
- Clear per-second API pricing and both sync and async endpoints
- Available through API, visual Studio, public weights, and ComfyUI
- Current 2.5 generation improves multi-shot consistency and detail
What to consider
- The original ltx-2-fast model ID has been retired and must be replaced in integrations
- Fast prioritizes speed and cost; LTX-2.5 Pro may produce higher fidelity for supported resolutions
- Retake, extend, and reframe endpoints currently require LTX-2.3 Pro rather than LTX-2.5 Fast
- Maximum duration falls from 20 to 10 seconds at high frame rates or resolutions above 1080p
- Automatic-duration prepaid requests require enough balance to cover the longest possible output before unused credit is released
- Studio credits and API credits are separate billing systems
- Free and Lite Studio plans are restricted to personal use
- Generated people, dialogue, typography, motion, and continuity still require careful review
Compare
LTX-2.5 Fast (successor to LTX-2-Fast) alternatives
The right alternative depends on the specific output, workflow, controls and budget your project requires.
Content Creator
Runway Gen-4.5
Choose Runway Gen-4.5 when an established hosted editing and generation suite is more important than open-weight deployment.
Explore Runway Gen-4.5 →Content Creator
Wan 3.0
Choose Wan 3.0 when longer clips, multimodal reference assets, and Alibaba Cloud's production API fit the workflow better.
Explore Wan 3.0 →Content Creator
Hailuo AI
Choose Hailuo AI when MiniMax's consumer-friendly video generation workflow and its visual style are a better match.
Explore Hailuo AI →Questions
LTX-2.5 Fast (successor to LTX-2-Fast) FAQs
Is LTX-2-Fast still available?
No. The original ltx-2-fast API model was retired in August 2026. Migrate to ltx-2-5-fast, ltx-2-5-pro, or a supported LTX-2.3 model.
What is the replacement for LTX-2-Fast?
LTX-2.5 Fast is the current speed-focused successor. It supports text-to-video, image-to-video, and audio-to-video in portrait or landscape up to 4K.
How much does LTX-2.5 Fast cost through the API?
Current prices are $0.09 per second at 720p, $0.13 at 1080p, $0.19 at 1440p, and $0.30 at 4K.
How long can LTX-2.5 Fast videos be?
Text and image generations can reach 20 seconds at 720p or 1080p with 24 or 25 fps. Higher frame rates and 1440p or 4K outputs are limited to 10 seconds.
Can I run LTX models locally?
Yes. Lightricks publishes inference code and gated model weights, plus ComfyUI and training workflows. You must supply compatible hardware and follow the model's license terms.
Bottom line
Our LTX-2.5 Fast (successor to LTX-2-Fast) verdict
LTX-2.5 Fast is the right upgrade for most former LTX-2-Fast users who value rapid synchronized audio-video generation, a broad resolution range, and transparent API costs. Test it at 720p or 1080p before scaling, and choose Pro or another model when fidelity, editing endpoints, longer duration, or production consistency outweigh speed.
Visit LTX-2.5 Fast (successor to LTX-2-Fast) website ↗Similar Content Creator Tools
View all Content Creator AI tools →
Sonic 3
Sonic 3 - Cartesia's realistic text-to-speech model for voice agents
Canva Creative Operating System
Canva Creative Operating System - A supercharged Visual Suite with a design-focused AI model, video tools, marketing upgrades, and more.
Hailuo 2.3
Hailuo 2.3 - MiniMax's new AI video model with upgraded movement, realism, and expression
Inworld Realtime TTS
Inworld TTS - Voice AI with cloning, multilingual support, real-time streaming, and emotion
AI training for the future of work.
Get access to all our AI courses, hundreds of real-world AI use cases, live expert-led workshops, an exclusive network of AI early adopters, and more.
AI Courses
Get unlimited access to all of our current & upcoming industry-specific AI courses for the duration of your subscription.
Daily Guides
To keep up with the rapid pace of AI, our team publishes AI implementation guides daily. Our library contains 300+ practical use cases to automate real-world work.
Workshops
Join weekly, live, interactive sessions with industry leaders who are at the forefront of AI for hands-on implementation guidance and exclusive insights.
Community
Network with an exclusive community of AI-first professionals who are working smarter with AI. Learn how early adopters are using AI in their work and businesses.