Skip to content

v0.1.0: Agentic Visual Intelligence Alpha

Pre-release
Pre-release

Choose a tag to compare

@apundhir apundhir released this 31 Jan 07:31
· 16 commits to main since this release

Changelog

All notable changes to this project will be documented in this file.

[0.1.0] - 2024-01-31

🚀 Features

  • Visual Intelligence: Integrated Qwen2-VL (via Fireworks AI) for unified layout analysis and OCR.
  • Frontend:
    • Next.js 14 UI with Tailwind CSS and Shadcn/UI integration.
    • Interactive Document Visualizer with bounding box overlays.
    • Drag-and-drop file upload supporting PDF and Image formats.
  • Microservices Architecture:
    • Orchestrator: Asyncio-based workflow engine managed by FastAPI.
    • Preprocessing: PDF-to-Image conversion and normalization pipeline.
    • Visual Service: Dedicated service for VLM inference and JSON parsing.

🐛 Bug Fixes

  • Fixed "Image failed to load" race condition for multi-page PDF uploads.
  • Resolved race conditions in concurrent page processing.
  • Fixed TypeScript type errors in frontend components.

🛠 Improvements

  • Establishing comprehensive CI pipeline (Linting + Build Checks).
  • Standardized logging across all Python services.
  • Consolidated repository structure (removed nested git).
  • Added ARCHITECTURE.md with detailed Mermaid system diagrams.