What's New

Updated Vision Models

Google Gemini

• Added Gemini 3 Pro and Gemini 3 Flash (preview)
• Default changed to Gemini 3 Flash
• Removed deprecated Gemini 2.0 models

Anthropic Claude

• Updated to Claude 4.5 Sonnet and Claude 4.5 Haiku

OpenAI

• Added GPT-5 and GPT-5 Mini

PDF Page Range Selection

• Dual-slider UI for selecting which pages to process
• Privacy-preserving: Local uploads extract only selected pages client-side (unselected pages never leave your device)
• URL-based PDFs use Mistral's pages API parameter

Security & Maintenance

• Fixed npm vulnerabilities (axios, glob, js-yaml, node-forge, etc.)
• Added pdf-lib for client-side PDF manipulation