wjh1547485653-max/minicpm-vision-mcp

View on GitHub ↗

为 DeepSeek V4.0 等单模态大模型装上眼睛和耳朵 —— 本地视觉+音频 MCP 服务器,基于 Ollama + MiniCPM-V 4.6 + faster-whisper,图片描述·视频分析·语音转文字

13 ★2 forksJavaScriptUpdated 2mo ago

Topics

ai-agentdeepseekimage-descriptionlocal-aimcpminicpmmodel-context-protocolollamavideo-analysisvision
Stars
13★
Forks
2
Language
JavaScript
License
MIT
Created
2026-06-26
Last push
2026-06-26