LTX Video 2.3 Full Quality
fal-ai/ltx-2.3/text-to-videoltx-2-v2-3-full-text-to-video- 🎬 Open-source text-to-video model from Lightricks, full-quality tier.
- 🆕 LTX-2.3 update: improved audio, visual quality, and prompt adherence.
- 🎵 Generates synchronized video and audio from a single network.
- 📏 Supports extended clips and high-resolution output.
- 🔧 Updated VAE for finer detail over LTX-2.
- 🔒 Open weights released on Hugging Face under Apache 2.0.
- ⚡ Quantized and distilled variants plus upscalers for efficiency.
- 🌐 Diffusers integration and local execution supported.
Lightricks is a creative technology company founded in January 2013, widely recognized for its popular selfie-editing app Facetune. Originally focused on mobile video and image editing tools, the company has expanded significantly into generative AI, developing its own video…
Explore 5 more models by Lightricks →LTX Video 2.3 Full Quality is the high-fidelity, text-to-video tier of Lightricks' LTX-2.3 release, a diffusion-transformer foundation model that generates synchronized video and audio from a single network. It targets stable, detailed output for production-grade workflows, contrasting with the speed-focused LTX Video 2.3 Fast flow built for rapid iteration. An image-to-video counterpart in the same Full Quality tier, LTX Video 2.3 Full Quality, shares the same engine.
Compared with its same-family predecessor, LTX Video 2.0 19B, Lightricks documents LTX-2.3 as an update with improved audio and visual quality and enhanced prompt adherence. The release adds an updated VAE for finer detail and supports extended clip lengths at high resolution.
The model retains LTX-2's core design philosophy: open weights, practical local execution, and one model handling both video and synchronized native audio. The release also provides quantized and distilled variants for faster inference, alongside spatial and temporal upscalers that let creators render at manageable resolution then scale up.
Weights are freely available on Hugging Face under the Apache 2.0 license, with reference workflows documented on the model card. It integrates with the Diffusers library, fitting the family's emphasis on iterative, exploratory video creation.
This About section is AI-generated from public sources via VeniceStats + Venice inference, with no human editing. It may contain inaccuracies.
| Seller | Rep↓ | Sales | Video | Routing | $ / s | $ / video |
|---|---|---|---|---|---|---|
| ▲ Apex Ant 0x73b4…e736 | 81.71 | — | — | #1 | $0.0394 | — |
| antseed-tidal-falcon-88cd 0x80d1…88cd | 12.78 | — | — | gated | $0.092 | — |
"Best price" and the seller table are live Antseed catalog data (advertised $/1M tokens — or $ per generated image for unit-billed image models — not settled amounts). Reputation = buyer trust score (0-100, the Antseed SDK's own formula). "Routing" = the SDK's default buyer routing (what the VPR desktop app ships with): a trust ≥ 60 gate on the effective reputation, then cheapest-first among routable sellers; live failover state (per-peer cooldowns) is buyer-side runtime and not included. Model knowledge (TLDR, provider, About) via the VeniceStats enrichment layer. Advertised catalog, not the model used in any specific purchase. "Usage on Antseed" counts only settlements whose buyers share the per-model split on-chain (metadata v2/v3, opt-in), so every usage figure is a lower bound.