MiniCPM-V-4.7-1B

What is MiniCPM-V-4.7-1B

MiniCPM-V-4.7-1B is an open-source multimodal vision-language model released by OpenBMB, with only 1B parameters. It supports image understanding and multimodal dialogue.

Key features

• Supports image understanding and multimodal dialogue
• Uses intra-ViT early compression technology to compress visual tokens early within the vision encoder
• Supports hybrid 4x/16x visual token compression, balancing visual detail with inference cost
• Supports on-device deployment on iOS, Android and HarmonyOS
• Provides out-of-the-box support for chip platforms including Rockchip, Houmo, Intel, AMD and Arm

Who it is for

Suitable for developers needing image understanding and multimodal dialogue on mobile or edge devices, and for multimodal applications seeking high performance density and efficient inference.

How to get started

1. Visit https://huggingface.co/openbmb/MiniCPM-V-4.7-1B to download the model weights
2. Refer to the Usage section in the model card for inference and deployment
3. Use the provided open-source on-device deployment support to run on iOS, Android or HarmonyOS

Compiled from public descriptions on the tool's official website / README / launch page. Pricing and user numbers have not been verified; please check the official website. Some sections were drafted with AI assistance. Written in Chinese and English; other languages are machine-translated.