Iris-3B: open-weights 3B pixel-space diffusion T2I model with no VAE, Apache 2.0
_akhaliq · x · 2026-10-09
Sperid Labs released Iris-3B, a 3B-parameter text-to-image model trained from scratch entirely in pixel space — generating every pixel directly with no VAE. The paper also explores its generative prior for detail-critical tasks like depth estimation and image restoration. Weights, code, and demo are fully open under Apache 2.0 on Hugging Face.
Related event: Sperid Labs Open-Sources Pixel-Space Text-to-Image Model Iris-3B(4 posts)→
More from Multimodal
- Speridlabs Releases Iris-3B: Pixel-Space Generative Model That Can Replace DINOv2 — Apprehensive_Sky892 · 2026-10-09
- GPT-6 Luna Shows Surprising Skill at Spotting AI-Generated Images — Angaisb_ · 2026-10-09
- KAIST's LongTake: long-horizon teacher forcing keeps 30-60s video generation dynamic — kaist-ai · 2026-10-09
- AI image of the day: GPT-6 Image 2.5 recreates 1940s Chinese village newsreel — DeryaTR_ · 2026-10-09
- Video character swap on 8GB VRAM: 640x480 render in 10:33 on RTX 4070 — big-boss_97 · 2026-10-09
- Google opens SynthID Detector globally; Vidu Q4 Preview prices video gen from $0.013/sec — 创业邦 · 2026-10-09