LTX Video 2.5 Fast
LTX Video 2.5 Fast is an open-weights, high-speed AI video model for text-to-video and image-to-video generation, optimized for rapid iteration and production workflows.
Overview
What is LTX Video 2.5 Fast
LTX Video 2.5 Fast is an open-weights AI model by Lightricks that generates up to 20-second videos from text or images. Released in August 2026, it supports resolutions up to 2160p, audio integration, and fast rendering — ideal for storyboards, previews, and scalable video production.
Running it privately on Venice
On Venice, LTX Video 2.5 Fast runs with anonymized privacy — your prompts are never stored or used for training. This ensures full creative sovereignty while accessing open, uncensored video generation. Run both text-to-video and image-to-video variants permissionlessly, with end-to-end control over your assets.
Assessment
Strengths and limitations
- Extremely fast generation: produces a 10-second clip in under 7 seconds on NVIDIA superchips.
- Open weights under Apache 2.0 license: free to use, modify, and deploy commercially.
- Supports both text-to-video and image-to-video workflows within the same model family.
- Native audio generation synchronized with visuals, enhancing production readiness.
- Available in high resolutions up to 4K with support for portrait and landscape aspect ratios.
- Lower motion stability and detail fidelity compared to the higher-tier LTX-2.5 Pro.
- Limited character consistency across scenes: better suited for single shots than multi-shot narratives.
- No local fine-tuning support in the Fast variant; only available via API or cloud inference.
Samples
Sample outputs
Generated on Venice with our standard prompt suite — the same prompts we run through every model of this type, so you can judge it like-for-like.
Slow aerial drone shot gliding over a misty mountain valley at golden hour, sunlight piercing clouds onto a winding river, ancient pine forests on either side, ultra-smooth motion, professional color grading, atmospheric haze, 4K
Calm ocean waves rolling onto a black-sand beach at sunrise, golden light on wet sand, foam dissolving into the shore, a single silhouetted figure at the waterline, smooth continuous forward push, serene cinematic atmosphere
A slow tracking shot through a rain-soaked Tokyo street at night, neon reflecting in puddles, steam rising from a food cart, a person with a translucent umbrella, shallow depth of field, teal-and-orange grade, smooth steady camera
Capabilities
What it supports
- Text to video
- Image to video
- Reference to video
- Native audio generation
Variants
LTX Video 2.5 Fast model variants
LTX Video 2.5 Fast runs on Venice as 2 variants of the same underlying model. Pick by what you're starting from: a written prompt, a still image, reference images, or an existing clip. Each variant is its own model id on the API; the generation quality is the same across the family.
| Variant | What it is | Clip lengths | Resolutions | Aspect ratios | Audio | Model ID |
|---|---|---|---|---|---|---|
| Text to Video | Generate a clip from a written prompt | 6s – 20s | 720p, 1080p, 1440p, 2160p | 16:9, 9:16 | ltx-2-5-fast-text-to-video | |
| Image to Videoflagship | Animate a still image into motion | 6s – 20s | 720p, 1080p, 1440p, 2160p | auto, 16:9, 9:16 | ltx-2-5-fast-image-to-video |
Capability data comes straight from the Venice model API and refreshes with every catalog ingest. The specs and pricing on this page are captured from the flagship variant; pass the model id of the variant you want to the API.
LTX Video 2.5 Fast Text to Video
Generate a clip from a written prompt. Supports clips of 6s – 20s, 720p, 1080p, 1440p, 2160p output, 16:9, 9:16 aspect ratios, with native audio.
ltx-2-5-fast-text-to-videoLTX Video 2.5 Fast Image to Video
Animate a still image into motion. Supports clips of 6s – 20s, 720p, 1080p, 1440p, 2160p output, auto, 16:9, 9:16 aspect ratios, with native audio.
ltx-2-5-fast-image-to-videoSpecifications
Datasheet
- Maker
- Lightricks
- Released
- August 2026
- Modality
- Text-to-video, image-to-video
- Max resolution
- 2160p (4K)
- Resolutions
- 720p, 1080p, 1440p, 2160p
- Clip lengths
- 6s – 20s in 2s steps
- Mode
- image-to-video
- Aspect ratios
- auto, 16:9, 9:16
- Audio
- Yes
- Privacy on Venice
- Anonymized — prompts not stored
- Available on Venice since
- Aug 2026
- License
- Apache 2.0
API
Call it from your code
Venice exposes this model through the REST API. Queue a generation with the model id.
curl https://api.venice.ai/api/v1/video/queue \
-H "Authorization: Bearer $VENICE_API_KEY" \
-H "Content-Type: application/json" \
-d '{
"model": "ltx-2-5-fast-image-to-video",
"prompt": "Aerial drone shot over a misty mountain valley at golden hour"
}'
# Use the returned queue_id with https://api.venice.ai/api/v1/video/retrieve.
# Call /video/complete after downloading if needed.Pricing
What it costs on Venice
Pay per clip on Venice — price scales with resolution and duration (6s–20s), from $0.59.
New Venice accounts include a free daily allowance and 500 welcome credits — no credit card required.
Alternatives
How it compares
| Model | Max resolution | Strongest at | Open weights | Price (Venice) |
|---|---|---|---|---|
| LTX Video 2.5 Fast | 2160p | Speed & open access | Yes | from $0.59 |
| Wan 2.7 | 1080p | Stylized motion | Yes | from $0.55 |
| Kling O3 Pro | 1080p | Cinematic realism | No | from $0.46 |
| Vidu Q3 | 1440p | Longer clips | No | from $0.27 |
Fastest open-weights video model with full API control and no usage caps.
Use cases
What it is good for
- 01Rapid prototyping and storyboarding for film or advertising.
- 02Social media content creation with vertical and horizontal formats.
- 03Animating still images into engaging clips for marketing or education.
- 04Audio-visual demos where sound and motion are tightly synchronized.
- 05Cost-sensitive, high-volume video generation for A/B testing or personalization.
Prompting
Getting better results
Use clear, concise prompts focusing on action and camera movement for best results.
Start with 720p or 1080p for faster iterations before scaling to 1440p or 2160p.
Leverage automatic duration prediction by describing the required action rather than specifying time.
Version history
Initial release, deprecated as of August 2026.
Major upgrade with 4K and audio support.
Current — optimized for speed and open deployment.
FAQ
Frequently asked questions
LTX Video 2.5 Fast is an open-weights AI video generation model by Lightricks, designed for fast, scalable production of up to 20-second clips from text or images. It supports resolutions up to 4K and includes synchronized audio output.
On Venice, pricing starts at $0.59 per clip for 720p at 6 seconds, scaling with resolution and duration. You pay per generation with no subscription, making it cost-effective for prototyping and high-volume use.
Yes. LTX Video 2.5 Fast is open weights under the Apache 2.0 license, meaning it's free to use, modify, and deploy commercially. The model weights are publicly available.
Yes. The variant 'ltx-2-5-fast-image-to-video' specifically enables animating still images into video clips up to 20 seconds long, with support for multiple aspect ratios and resolutions.
It supports 720p, 1080p, 1440p, and 2160p (4K) resolutions, with options for landscape (16:9) and portrait (9:16) aspect ratios depending on use case.
Yes. It generates synchronized audio alongside video, making it suitable for production workflows that require sound design, voiceovers, or music timing.
LTX Video 2.5 Fast wins on resolution, speed, and production features like audio sync and aspect flexibility. Wan 2.7 is open weights too but limited to 1080p and less optimized for commercial pipelines.
No. LTX Video 2.5 Fast is a generative video model and does not support external tool use or function calling. It focuses purely on media synthesis from prompts or images.
Yes. The 'ltx-2-5-fast-text-to-video' variant allows generating video directly from written prompts, supporting durations from 6 to 20 seconds across multiple resolutions and aspect ratios.
Run LTX Video 2.5 Fast privately
No prompt logging. No data used for training.