Private AI

Nano GPT logo
NanoGPT
Nano GPT logo
NanoGPT
Nano GPT logo
NanoGPT
Search
 
 
 
 
All conversations

Create

Chat
Studio

Explore

Models
Pricing
AI Apps
Blog

Library

Gallery
Workspace
Usage & billing
API
Suggestions and bugs
Use cases

|

Terms

|

Privacy

|

Refunds
Usage & billing
Teams
Referrals
Presets
Settings
FAQPress & mediaBrand assetsBenchmarksNano RPCEarn crypto
  1. Models
  2. Video

LTX-2.3 Quality

Try Model

LTX-2.3 Quality routes text, image, audio, reference video, extend-video, video-to-HDR, and optional LoRA inputs to the appropriate generation mode. Supports output sizes such as landscape, portrait, square, and auto with native synchronized audio.

Added Jun 2, 2026

Details

Starting Price
From $0.092 per video
Model type
both

Final price depends on the selected settings and is shown before generation.

Settings

Generation controls available for this model.

Output Format

landscape_16_9

Default Duration

6

9 duration options

Duration

Select

Default

6

Options (9)

5 seconds, 6 seconds, 8 seconds, 10 seconds +5 more

Clip length

Extend Direction

Select

Default

end

Options (2)

Forward, Backward

Choose whether to generate after the source video or before it.

Frames Per Second

Select

Default

24

Options (1)

24 FPS

Frames per second

Generate Audio

Toggle

Default

Yes

Enable synchronized audio when supported

Input Mode

Select

Default

auto

Options (3)

Auto, Extend Video, HDR

Auto picks the endpoint from uploaded media; Extend continues an uploaded source video.

LoRA 1 Scale

Number

Default

1

Strength of LoRA 1 (0.0-2.0).

LoRA 1 URL

Text

Default

N/A

Optional LoRA weights URL (max 3 total).

LoRA 2 Scale

Number

Default

1

Strength of LoRA 2 (0.0-2.0).

LoRA 2 URL

Text

Default

N/A

Optional second LoRA weights URL.

Showing 9 of 12 settings.

Benchmarks

No public benchmark scores for this model yet.

Examples

Loading examples…

Related video models

Compare LTX-2.3 Quality with similar models from the same provider or model family.

LTX-2 19B

ltx-2-19b

Unified LTX-2 19B model for text-to-video or image-to-video with synchronized audio. Supports optional LoRA adapters for custom styles.

Lightricks LTX-2 Fast

lightricks-ltx-2-fast

High-speed LTX-2 pipeline tuned for rapid iterations. Convert text or a single image into cinematic clips with synchronized audio in seconds.

Lightricks LTX-2 Pro

lightricks-ltx-2-pro

Flagship LTX-2 stack for production-ready motion. Generates synchronized audio and rich camera moves from text prompts or reference images.

LTX-2.5 AlphaGen

lightricks/ltx-2.5/alpha-gen

Extract a frame-matched foreground matte and a ProRes 4444 cutout with straight alpha from existing footage. Preserves soft edges, hair and partial transparency. Both outputs are silent. Accepts MP4 or MOV up to 200 MB and 3072 frames, with long side at most 3840 pixels, short side at most 2160 pixels and aspect ratio at most 32:9.

LTX-2.5 HDR

lightricks/ltx-2.5/hdr

Convert existing SDR footage into a 10-bit HEVC BT.2020/HLG master and downloadable 16-bit half-float OpenEXR frames. Choose ACEScg, ACEScct, linear sRGB or ACES2065-1 for the EXR export. Accepts MP4 or MOV up to 200 MB and 3840×2160 in either orientation, with limits of 273, 193 or 65 input frames at the 1080p, 1440p or 4K pixel tier. Assess the master in a compatible HDR player and display.

LTX-2.5 Reframe

lightricks/ltx-2.5/reframe

Expand an existing video into a new composition by generating the surrounding scene while keeping the whole source visible. Choose 720p or 1080p at 1:1, 4:5, 5:4, 9:16 or 16:9, with optional anchor or rectangle placement. Placement preserves source proportions and allows at most 2× enlargement. Accepts MP4 or MOV up to 200 MB, 60 seconds and 1800 frames; each input side must be 64–4096 pixels.