MosiAI Open-Sources MOSS-Image With Day-0 Support From vLLM-Omni

1 reports 1 reporting sources 3 hours agoNew

Understand the event

AI overview

MosiAI released MOSS-Image’s model weights and inference code, with Base and SFT checkpoints available on Hugging Face, while vLLM announced Day-0 support through vLLM-Omni. The inference code is licensed under Apache-2.0. Built on a 5B Diffusion Transformer (DiT), MOSS-Image supports Chinese and English text rendering and prompts of up to 4K tokens. It also offers experimental JSON layout control using bounding boxes.

Generated by AI from reports · Updated 2 hours ago

Latest development10/17 10:24 (UTC+8)

MOSS-Image goes open source with bilingual text rendering and experimental layout control vLLM · X

Report timeline

Follow the reports to explore different sides of the event.

10/17

10:24
vLLM · XPicks
MOSS-Image goes open source with bilingual text rendering and experimental layout control
MosiAI announced the open-source release of MOSS-Image, built on a 5B Diffusion Transformer (DiT), with Chinese and English text rendering, up to 4K-token prompts, and experimental JSON layout control using bounding boxes. Base and SFT checkpoints are available on Hugging Face, alongside open model weights and Apache-2.0 licensed inference code. vLLM congratulated the team and announced Day-0 support through vLLM-Omni.

Event heat trend

There is not enough continuous observation data to plot a trend yet.

Keep reading

Latest daily · Issue 24NVIDIA and AI4Bharat Introduce Nemotron-pretraining-Indic-v1 and an Open Data Generation Pipeline