Tencent's WeChat Team Releases 31.7B Unified Image Model WeLM-Image

Tencent's WeChat Vision team released WeLM-Image, a 31.7B-parameter unified image foundation model that integrates multimodal understanding into text-to-image generation, instruction-based editing, and multi-reference composition. It is designed for single-GPU deployment at WeChat's consumer scale, generating 1024px images in about 5 seconds.

2026-11-06 ~ 2026-11-08 · 2 related posts

• WeChat unveils WeLM-Image: unified 31.7B-param image model tops open-source benchmarks — Canmiao Fu · 2026-11-06
• Tencent unveils WeLM-Image: 31.7B MoE image model tops open-source benchmarks, 5s per image on one H20 — Crazy-Repeat-2006 · 2026-11-08