Alibaba's Qwen released Qwen-MM-Plugins on August 10, a multimodal plugin collection that packages vision, video, OCR, object grounding, segmentation, speech transcription, and 3D operations as callable tools for agent frameworks including Claude Code, Codex, Qwen Code, and Gemini CLI.
The release is separate from Agent Plugins, the open standard OpenAI and Vercel introduced on August 6 with AWS, Cursor, GitHub, and VS Code to allow a plugin built once to run across any compatible agent client.