ThirdBrAIn.tech

Install Cherry Studio with Ollama on Windows Desktop App for LLMs, RAG, MCP, Agents

2 min read

Install Cherry Studio with Ollama on Windows Desktop App for LLMs, RAG, MCP, Agents

AI Summary

This video demonstrates how to install and set up Cherry Studio, an open-source desktop LLM client, on Windows with Ollama integration.

Key Features of Cherry Studio

Multi-Provider Support: Cherry Studio supports numerous LLM providers including:

  • Ollama (local models)
  • OpenAI GPT models
  • Gemini
  • Claude (Anthropic)
  • DeepSeek
  • Many other providers

Core Capabilities:

  • LLM inference with multiple providers
  • RAG (Retrieval-Augmented Generation) functionality
  • Web searching integration
  • Multi-modal capabilities (text and image)
  • Pre-built agents for various tasks
  • Knowledge base management
  • MCP (Model Context Protocol) server support

Installation Process

System Requirements:

  • Windows (not Windows 7) or macOS
  • No Linux support currently
  • Lightweight application

Installation Steps:

  1. Visit Cherry Studio’s installation page
  2. Download the appropriate version (x64 for Windows)
  3. Run the installer with standard Next/Next process
  4. Launch the application after installation

Ollama Integration Setup

Prerequisites:

  • Install Ollama on your system
  • Download desired models (e.g., Qwen 3)
  • Ensure Ollama service is running

Configuration:

  1. In Cherry Studio settings, select Ollama provider
  2. Use default localhost:11434 endpoint (adjust if modified)
  3. Select downloaded models from the manage models section
  4. Models show capabilities (reasoning, tool calling, etc.)

Demonstrated Workflow

Hardware Setup: Video uses A10G GPU with 24GB VRAM for optimal performance

Basic Usage:

  1. Configure model provider (Ollama)
  2. Select specific model (Qwen 3)
  3. Start chatting with very fast response times
  4. Switch between different providers as needed

Advanced Features:

  • Agents: Pre-built agents available for download
  • Multimodality: Image generation and processing
  • Knowledge Base: Upload PDFs and documents for RAG
  • Translation: Built-in translation capabilities

Limitations and Considerations

Language Barrier: Most documentation is in Chinese, requiring translation for English users

Provider Dependencies: Some features require API keys for external providers

Platform Limitation: Currently only supports Windows and macOS

Comparison with Alternatives

Cherry Studio positions itself as a comprehensive solution compared to other tools like:

  • Jan AI
  • AnythingLLM
  • LM Studio

The key differentiator is its attempt to provide quality implementation across all features rather than excelling in just one area.

Conclusion

Cherry Studio appears to be a well-designed, lightweight desktop application that successfully integrates multiple LLM providers with advanced features like RAG and agent support. While the Chinese documentation presents a barrier, the interface is intuitive enough for English users to navigate effectively. The tool shows promise for users wanting a unified interface for various LLM capabilities without the complexity of command-line tools.

Graph View