khoj-ai/khoj 1.9.0

on GitHub

🎉 New

• Upgrade Offline Chat in Khoj to use Llama.cpp by @debanjum in #680
  • Upgrade default chat model to NouseResearch's Hermes-2-Pro
  • Support using any chat model in GGUF format
  • Support more GPUs like AMD, Nvidia, Mac, Vulcan. Previously we just supported Vulcan, Mac
• Stream Khoj's intermediate actions during chat to the Web UI by @sabaimran in #679

🧪 Improve

• Extract More Dates from entries to improve Date Filter by @debanjum in #683

⚙️ Fix

• Address Notion, Image tech debt in indexing code path by @sabaimran in #687

Full Changelog: 1.8.0...1.9.0