Text-to-Video Image-to-Video Audio-to-Video

LTX-2.3 22B

Free Playground & API · $5 to Start

State-of-the-art video generation model with 22 billion parameters. Supports text-to-video, image-to-video, and audio-to-video with high-quality cinematic output.

No subscription
No credit card required

Sample Outputs

Real outputs generated with LTX-2.3 22B on deAPI.

SHOT / 8
24 121 768×512
24 121 768×512
24 121 768×512
24 121 768×512
24 121 768×512
24 121 768×512
24 121 768×512
24 121 768×512

"A medium close-up frames an elderly fisherman mending a net on a weathered wooden pier at dawn. His calloused hands work methodically, threading thick rope through torn sections. The first light of morning paints the scene in soft peach and lavender tones. The camera pushes in slowly, settling on his concentrated expression. Seagulls cry in the distance."

"A wide establishing shot captures a narrow Venice canal at blue hour. Ornate stone buildings line both sides, warm amber window lights reflecting in long shimmering streaks across the dark water. A single gondola glides silently from left to right, the gondolier pushing his oar in slow rhythmic strokes. Distant accordion music floats across the water."

"An extreme close-up reveals a monarch butterfly emerging from its chrysalis in slow motion. The translucent shell splits and a crumpled wet wing unfolds tentatively outward. Sunlight backlights the wing membrane, revealing an intricate network of orange and black veins. The camera holds on a macro lens, the garden behind dissolving into soft bokeh."

"A bird's eye view looks straight down onto a busy Tokyo intersection at night as pedestrian lights turn green. Hundreds of people cross from all four directions, their umbrellas forming a moving mosaic of black, red, and transparent circles against wet asphalt. Neon signs reflect in the rain-slicked road. The camera slowly rotates clockwise as the crowd flows."

"A low-angle medium shot captures a blacksmith hammering a glowing orange piece of metal against an iron anvil. Each strike sends bright sparks cascading outward in the darkness. The forge fire behind him pulses with deep red and orange light, illuminating the sweat on his forearms. Between strikes, he examines the blade edge, turning it slowly. The heavy clang echoes off stone walls."

"An aerial drone shot drifts forward over a winding mountain road in autumn, hillsides ablaze with red, orange, and gold foliage. A single white car traces the curves far below. Patches of morning fog cling to the lower ravines, slowly dissolving as sunlight reaches them. The camera maintains a smooth forward glide at medium altitude, the landscape unfolding continuously."

"A medium shot frames a young street musician playing violin on a rainy European bridge at twilight. Raindrops catch the amber glow of vintage streetlamps. She sways gently with the music, eyes closed, bow arm drawing long expressive strokes. The camera slowly orbits around her, transitioning from front view to three-quarter profile. The violin melody mixes with steady rain."

"A close-up captures a glass orb sitting on moss-covered forest floor, the surrounding woodland reflected upside-down in its curved surface. A ladybug crawls slowly across the top of the sphere. Morning dew drops cling to the moss, sparkling as a shaft of sunlight breaks through the canopy. Gentle birdsong and the creak of swaying trees provide the soundtrack."

Capabilities

Model ID: Ltx2_3_22B_Dist_INT8

What you can do with LTX-2.3 22B.

Text-to-Video

Create dynamic videos from text prompts. Generate engaging visual content without cameras or editing software.

Image-to-Video

Bring static images to life with AI-powered animation. Transform photos into short, captivating video clips.

Try it

Audio-to-Video

Generate lip-synced talking head videos from audio and a portrait image. Perfect for AI avatars, presentations, and content creation.

Technical Specifications

Resolution

512–1024 x 512–1024 px

Default Size

768 x 768 px

Inference Steps

8 steps (default: 8)

Guidance

Not required

Video Length

49–241 frames at 24 fps (2–10s)

How to Use in deAPI

Integrate LTX-2.3 22B into your app with a single API call.

POST https://api.deapi.ai/api/v1/client/txt2video
curl -X 'POST' \
  'https://api.deapi.ai/api/v1/client/txt2video' \
  -H 'accept: application/json' \
  -H 'Content-Type: application/json' \
  -H 'Authorization: Bearer YOUR_API_KEY' \
  -d '{
    "model": "Ltx2_3_22B_Dist_INT8",
    "prompt": "a cat walking through a garden",
    "width": 768,
    "height": 768,
    "num_frames": 65
}'
{
    "model": "Ltx2_3_22B_Dist_INT8",
    "prompt": "a cat walking through a garden",
    "width": 768,
    "height": 768,
    "num_frames": 65
}

Tip: The API returns a request_id. Use webhooks (recommended) or poll GET /request-status/{'{'}request_id{'}'} for results.

POST https://api.deapi.ai/api/v1/client/img2video
curl -X 'POST' \
  'https://api.deapi.ai/api/v1/client/img2video' \
  -H 'accept: application/json' \
  -H 'Content-Type: application/json' \
  -H 'Authorization: Bearer YOUR_API_KEY' \
  -d '{
    "model": "Ltx2_3_22B_Dist_INT8",
    "prompt": "gentle camera zoom, cinematic lighting",
    "image": "BASE64_OR_URL",
    "num_frames": 65
}'
{
    "model": "Ltx2_3_22B_Dist_INT8",
    "prompt": "gentle camera zoom, cinematic lighting",
    "image": "BASE64_OR_URL",
    "num_frames": 65
}

Tip: The API returns a request_id. Use webhooks (recommended) or poll GET /request-status/{'{'}request_id{'}'} for results.

POST https://api.deapi.ai/api/v1/client/aud2video
curl -X 'POST' \
  'https://api.deapi.ai/api/v1/client/aud2video' \
  -H 'accept: application/json' \
  -H 'Content-Type: application/json' \
  -H 'Authorization: Bearer YOUR_API_KEY' \
  -d '{
    "model": "Ltx2_3_22B_Dist_INT8",
    "audio": "AUDIO_URL_OR_BASE64",
    "image": "BASE64_OR_URL"
}'
{
    "model": "Ltx2_3_22B_Dist_INT8",
    "audio": "AUDIO_URL_OR_BASE64",
    "image": "BASE64_OR_URL"
}

Tip: The API returns a request_id. Use webhooks (recommended) or poll GET /request-status/{'{'}request_id{'}'} for results.

Need an API key?

Sign up for free and get $5 in credits to start.

Get $5 credits
Get started with LTX-2.3 22B

Frequently Asked Questions

State-of-the-art video generation model with 22 billion parameters. Supports text-to-video, image-to-video, and audio-to-video with high-quality cinematic output.
LTX-2.3 22B is the largest model in the LTX family with 22 billion parameters. It uniquely supports three modes: text-to-video, image-to-video, and audio-to-video (lip-sync). Resolution up to 1024×1024 at 24fps, 2–10 second clips.
Provide a portrait image and an audio clip (1–11 seconds). LTX-2.3 22B generates a video where the person in the image appears to speak the audio with synchronized lip movements. Ideal for AI avatars and talking head content.
Yes. In image-to-video mode, you can set both a first frame and a last frame. The model will generate a smooth transition between the two, giving you precise control over the animation's start and end states.
Sign up for a free deAPI account at https://deapi.ai/register and get $5 in free credits — no credit card required. Then open the playground, select LTX-2.3 22B, and generate your first result in seconds.
deAPI is pay-per-use with no subscriptions or monthly fees. Check the pricing page at https://deapi.ai/pricing for current per-generation rates.
Yes. LTX-2.3 22B is available via the deAPI REST API. Submit a request, receive a request ID, and poll for the result. See the API documentation at https://docs.deapi.ai for code examples in Python, Node.js, and cURL.

Try LTX-2.3 22B for Free

Get $5 in free credits and start generating with LTX-2.3 22B.

Get $5 credits
No subscription
No credit card required
Pay only for what you use