OpenMOSS releases open-weight text-to-image model MOSS-Image
OpenMOSS released MOSS-Image, an Apache 2.0 open-weight 5B DiT text-to-image model with Qwen3.5-9B text encoder, multilingual text rendering and JSON layout control; its API is free and it is available in ComfyUI and Maestro v3.1.
2026-10-20 ~ 2026-10-21 · 3 related posts
- Episode 1: Fudan's OpenMOSS Open-Sources 5B MOSS-Image Text-to-Image Models(2026-10-16, 14 posts)
- Episode 2: Distilled MOSS-Image Model Generates Images in 8 Steps, Trends on Hugging Face(2026-10-16, 2 posts)
- Episode 3: Ostris AI Toolkit adds support for MOSS Image text-to-image models(2026-10-16, 3 posts)
- Episode 4: Fudan's OpenMOSS releases MOSS-Image, a 5B open text-to-image model(2026-10-18, 4 posts)
- Episode 5: MOSS-Image passes 354K downloads on Hugging Face(2026-10-18, 2 posts)
- Episode 6: OpenMOSS releases open-weight text-to-image model MOSS-Image(2026-10-20, 3 posts)
- OpenMOSS releases Apache 2.0 MOSS-Image: 5B DiT with CJK text rendering and JSON layout control — crystal_alpine · 2026-10-20
- MOSS-Image API goes live for free with 4K-token prompts and text rendering — alexcovo_eth · 2026-10-20
- MOSS Image 5B lands in Maestro with raw JSON prompting for precise layout control — cocktailpeanut · 2026-10-21