Home
WeChat Vision Team Opensources General Multimodal Embedding Model WeMM-Embedding, Daily Call Volume Exceeds Ten Billion

Tencent’s WeChat Visual Team has officially open-sourced WeMM-Embedding, a versatile multimodal embedding model designed to support both text and image inputs. By leveraging an advanced technical architecture and refined training strategies, this model provides robust support for AI developers, fostering innovative applications of artificial intelligence across diverse industry verticals.
Available in three distinct sizes—2B, 4B, and 9B—WeMM-Embedding caters to a wide range of deployment requirements. Its unified architecture integrates two-stage training and knowledge distillation, significantly boosting multimodal comprehension while maintaining an optimal balance between performance and deployment efficiency.
Proven through rigorous testing in real-world business environments, the model is already integrated into WeChat’s recommendation and search systems, handling billions of daily requests. This full open-source release underscores WeChat’s commitment to an open multimodal AI ecosystem and offers substantial technical support for AI development across the industry.
Related article
OpenAI to Launch Lower-Cost ChatGPT Version, May Introduce Ad-Based Model to Compete with Netflix
OpenAI is set to broaden its audience by overhauling its current service structure and launching a more affordable ChatGPT subscription tier. Internal sources indicate the company is drawing inspiration from Netflix’s business model, aiming to bridge
Man Found Guilty in $8M AI Music Royalties Fraud
On March 24, 2026, the United States Attorney's Office for the Southern District of New York announced that Michael Smith, a man from North Carolina, had pleaded guilty to conspiracy to commit wire fraud. He used artificial intelligence technology to
Yuandao Fudao Unveils AI Big Reading, an Intelligent Reading Product for Adolescents
At the 2026 World Artificial Intelligence Conference (WAIC), Yuanfudao Group unveiled "Yuanfu AI Big Reading," a proprietary tool designed to transform reading from passive consumption into an active, AI-driven dialogue. By leveraging Socratic questi
Related Special Topic Recommendations
Comments (0)
0/500

Tencent’s WeChat Visual Team has officially open-sourced WeMM-Embedding, a versatile multimodal embedding model designed to support both text and image inputs. By leveraging an advanced technical architecture and refined training strategies, this model provides robust support for AI developers, fostering innovative applications of artificial intelligence across diverse industry verticals.
Available in three distinct sizes—2B, 4B, and 9B—WeMM-Embedding caters to a wide range of deployment requirements. Its unified architecture integrates two-stage training and knowledge distillation, significantly boosting multimodal comprehension while maintaining an optimal balance between performance and deployment efficiency.
Proven through rigorous testing in real-world business environments, the model is already integrated into WeChat’s recommendation and search systems, handling billions of daily requests. This full open-source release underscores WeChat’s commitment to an open multimodal AI ecosystem and offers substantial technical support for AI development across the industry.
OpenAI to Launch Lower-Cost ChatGPT Version, May Introduce Ad-Based Model to Compete with Netflix
OpenAI is set to broaden its audience by overhauling its current service structure and launching a more affordable ChatGPT subscription tier. Internal sources indicate the company is drawing inspiration from Netflix’s business model, aiming to bridge
Man Found Guilty in $8M AI Music Royalties Fraud
On March 24, 2026, the United States Attorney's Office for the Southern District of New York announced that Michael Smith, a man from North Carolina, had pleaded guilty to conspiracy to commit wire fraud. He used artificial intelligence technology to
Yuandao Fudao Unveils AI Big Reading, an Intelligent Reading Product for Adolescents
At the 2026 World Artificial Intelligence Conference (WAIC), Yuanfudao Group unveiled "Yuanfu AI Big Reading," a proprietary tool designed to transform reading from passive consumption into an active, AI-driven dialogue. By leveraging Socratic questi











