Mini toolsMiniCPM V4.6.
By OpenBMB
MiniCPM-V-4.6 is an Apache 2.0 image-text-to-text model from OpenBMB, described as a pocket-sized MLLM for ultra-efficient image and video understanding on phones. It is built on SigLIP2-400M and Qwen3.5-0.8B, inheriting MiniCPM-V’s single-image, multi-image, and video understanding capabilities while improving computation efficiency. The model introduces mixed 4x/16x visual token compression, reduces visual encoding FLOPs by more than 50%, supports deployment on iOS, Android, and HarmonyOS, and is adapted for Transformers, vLLM, SGLang, llama.cpp, Ollama, GGUF, BNB, AWQ, and GPTQ workflows.
New
Multimodal
Gen 3
Overview
MiniCPM-V-4.6 is OpenBMB’s open-source lightweight multimodal model for efficient image, multi-image, and video understanding on mobile and edge devices.
About OpenBMB
OpenBMB is short for Open Lab for Big Model Base. The goal of OpenBMB is to build the model base and toolkit for large-scale pre-trained language models.
View Company Profile
Last updated: May 22, 2026
