DeepSeek
Integrates DeepSeek's language models with Model Context Protocol-compatible applications.
关于
DeepSeek is a Model Context Protocol (MCP) server designed to seamlessly integrate DeepSeek's powerful language models, such as deepseek-reasoner and deepseek-chat, with MCP-compatible applications like Claude Desktop. It allows for anonymous API usage through a proxy and offers features like automatic model fallback, custom model selection, temperature control, token limit adjustment, and multi-turn conversation support. The server intelligently maps natural language requests to configuration changes and supports resource discovery for available models and configurations. It is particularly useful for training dialogue models and managing complex interactions that require context from previous messages.
主要功能
- Supports multi-turn conversations with preserved context and settings.
- Automatically falls back to an alternative model if the primary model is unavailable.
- Allows custom model selection, temperature control, and token limit adjustments.
- Provides resource discovery for available models and configurations.
- Integrates with MCP Inspector for local testing and debugging.
- 2 GitHub stars
使用案例
- Training and fine-tuning DeepSeek language models using properly formatted conversation data.
- Managing complex, multi-step interactions where context is crucial.
- Seamless integration of DeepSeek models with applications like Claude Desktop.
