MiniCPM-o 4.5 — Overview

The omni-modal flagship of the MiniCPM-o series — vision, speech, and full-duplex live streaming, all in a single model.

Highlights

• End-to-end omnimodal — vision encoder + speech encoder + LLM + TTS densely connected via hidden states
• Full-duplex live streaming — hear, see, and speak simultaneously
• Voice cloning and configurable assistant voice
• HuggingFace: openbmb/MiniCPM-o-4_5 · ModelScope: OpenBMB/MiniCPM-o-4_5

For real-time, full-duplex demos see the MiniCPM-o-Demo repository — it ships a turn-key web demo for the o-series with PyTorch + CUDA inference and a frontend.

Where to start

• Audio recipes: Speech-to-Text, Text-to-Speech, Voice Cloning
• Deploy: vLLM, SGLang, llama.cpp, Ollama
• Quantize: GGUF