NVFP4 Version of LTX-2.5 Text Encoder Released for ComfyUI
superPussyman · reddit · 2026-08-22
An NVFP4 quantized version of the LTX-2.5 Gemma-4 12B text encoder has been released for ComfyUI. The developer quantized the heavy decoder linear layers to native NVFP4 while maintaining original precision for embeddings, norms, and vision components. Testing indicates lower VRAM usage with no obvious quality degradation, successfully validated on Blackwell architecture.
More from Multimodal
- PrunaAI's P-Video-2 Pro adds image/video/audio references from $0.01/s — aziz4ai · 2026-10-16
- ElevenLabs blind tests Eleven v4 voice model, wins all 8 languages on expressiveness — JenniferHli · 2026-10-16
- HyperFrames adds Reference Any Video and auto Brand Kits for AI video agents — alifcoder · 2026-10-16
- 'Alien Metamorphosis': AI short made with Grok Imagine + Topaz — creatoroff · 2026-10-16
- Early local test of distilled FLUX.3 Klein: mediocre t2i, editing hopes remain — Ant_6431 · 2026-10-16
- Krea 3 vs Qwen 2.1 vs ZIT: better facial nuance, but unsettlingly realistic outputs — Crazy-Repeat-2006 · 2026-10-16