Snap-Solver  by Zippland

AI screenshot assessment tool for academic problems

Created 1 year ago
255 stars

Top 98.8% on SourcePulse

GitHubView on GitHub
Project Summary

Summary

Snap-Solver is an AI-powered desktop application designed to automate the process of solving academic and technical problems presented on a user's screen. Targeting students, test-takers, and self-learners, it simplifies online assessments and enhances knowledge acquisition by capturing any on-screen question via a hotkey and providing detailed, AI-generated solutions.

How It Works

The project utilizes a Flask web service architecture. A user interface interacts with a backend that orchestrates screenshot capture, Optical Character Recognition (OCR) via Mathpix for precise text and complex mathematical formula extraction, and AI analysis. It supports a diverse range of large language models, including offerings from OpenAI, Anthropic, DeepSeek, Alibaba, Google, and Doubao, enabling users to leverage varied AI capabilities. Real-time communication via Socket.IO facilitates cross-device collaboration, allowing seamless viewing and analysis of captured problems on mobile devices.

Quick Start & Requirements

  • Installation: Execute python app.py.
  • Prerequisites: Python 3.x. Requires API keys for at least one AI provider (OpenAI, Anthropic, DeepSeek, Alibaba, Google) and a Mathpix API key for optimal OCR.
  • Access: Available locally via http://localhost:5000 or on the local network via http://[computer_IP]:5000.
  • Links: No direct quick-start or demo links are provided.

Highlighted Details

  • Cross-Device Synergy: Capture problems on a desktop and access solutions on mobile devices.
  • Extensive AI Model Support: Integrates with major AI families like GPT, Claude, DeepSeek, Qwen, and Gemini.
  • Advanced OCR: Leverages Mathpix for accurate recognition of text and intricate mathematical notation.
  • Platform Agnostic: Compatible with Windows, macOS, Linux desktops, and accessible via mobile browsers.
  • Configurable AI: Features adjustable parameters for AI's thinking depth, temperature, and custom system prompts.

Maintenance & Community

  • Support: Paid deployment services are available via zylanjian@outlook.com.
  • Feedback: Bug reports and feature requests are managed through GitHub Issues.

Licensing & Compatibility

  • License: Apache 2.0.
  • Compatibility: The Apache 2.0 license permits commercial use and integration into proprietary software.

Limitations & Caveats

Relies heavily on users obtaining and configuring multiple external API keys for AI and OCR services. Performance and accuracy are contingent on screenshot clarity and the reliability of third-party API providers. Requires stable network connectivity for both API access and local network functionality.

Health Check
Last Commit

1 month ago

Responsiveness

Inactive

Pull Requests (30d)
0
Issues (30d)
0
Star History
11 stars in the last 30 days

Explore Similar Projects

Feedback? Help us improve.