ltxv2-video

1# LTX-2 / LTX-2.3 Video Workflows
2 
3## Version naming (read this first)
4 
5There is no "LTX 3.2" or "LTX2.3" as separate products. The user's shorthand refers to **Lightricks LTX-2.3**, a point release of the LTX-2 family. The lineage is:
6 
7- **LTX-Video** (2024): first text-to-video model from Lightricks.
8- **LTX-2 / LTX-V2** (Oct 2025): 19B-class DiT audio-video foundation model. Bundled checkpoint `ltx-2-19b-distilled.safetensors`, Gemma 3 12B text encoder.
9- **LTX-2.3** (released ~March 2026): **22B**-parameter DiT update. Rebuilt VAE (sharper textures/faces/hair/text), ~4x larger text connector (text projection) for prompt adherence, native 9:16 portrait, LoRA support, HiFi-GAN vocoder for cleaner synchronized audio, up to 4K@50fps / ~20s clips. Apache 2.0. **Distributed primarily as GGUF UNets** (community quants) plus separate VAE / text-encoder / text-projection files, NOT a single bundled checkpoint like LTX-2.
10 
11When the user says "LTX3.2" / "LTX2.3", treat it as **LTX-2.3**. This skill covers both LTX-2 (bundled checkpoint path) and LTX-2.3 (GGUF UNet path).
12 
13---
14 
15## ⭐ Render-verified correct setup (read this FIRST — 2026-06-19)
16 
17> The GGUF-UNet + `DualCLIPLoader` + `gemma_3_12B_it_fp4_mixed` path documented later
18> in this skill (the Aitrepreneur installer path) **produces soft/mushy video with
19> inaccurate faces and eyes.** It runs, but it is NOT the quality path. The setup
20> below is the official Comfy-Org template, render-proven sharp (1280×704, accurate
21…