WeLM-Image release
The story so far
Hugging Face has released the unified image generation model WeLM-Image for visual intelligence, integrating multimodal understanding to support text-to-image generation, instruction-guided editing, and multi-reference synthesis. The model is designed for WeChat's large-scale consumer applications on a single GPU, using a Mixture-of-Transformers (MoT) backbone and Mixture-of-Experts (MoE) modules, with 31.7B storage parameters and approximately 5.4B generation parameters. Its data engine combines knowledge-based visual vocabulary and large-scale visual clustering to balance semantic coverage and editing tasks.
AI-written from the reports · Updated 13 days ago
Coverage timeline
Follow the coverage to see every side of the story.
- Hugging Face Daily PapersWeLM-Image: A Unified Image Generation Model towards Visual Intelligence
WeLM-Image is a unified image base model that integrates multimodal understanding, supporting text-to-image generation, instruction-guided editing, and multi-reference synthesis.