OpenVision

소개

OpenVision is a server that leverages OpenRouter's vision models to provide image analysis capabilities within the Model Context Protocol (MCP) ecosystem. It allows AI assistants to analyze images by accepting various image input types like URLs, base64 encoded strings and local file paths and returns detailed descriptions and analyses based on user queries. Powered by flexible configuration and integrations with tools like Claude Desktop and Cursor, it offers a simple interface for enriching AI interactions with visual context.

주요 기능

• Accepts image inputs as URLs, base64 encoded strings, and local file paths.
• Provides detailed image analysis based on user queries and system prompts.
• Integrates seamlessly with Claude Desktop and Cursor via MCP.
• Supports multiple OpenRouter vision models, including qwen, Claude, and GPT-4o.
• 2 GitHub stars
• Easily configurable via environment variables for OpenRouter API key and model selection.

사용 사례

• Enabling AI assistants to understand and respond to visual content.
• Content moderation by identifying objects and scenes in images.
• Automated image analysis for data extraction and insights.