More UI Improvements

Now, in the main chat UI we added some much desired UI improvements and fixes.

• New prompt input
• Better Citations UI and reporting
• Metrics for Agent calls
• Report document and web-search citations during Agent calls!
• Ability to each toggle on/off Agent skills from the prompt
• Ability to select the provider and model for the workspace without leaving the page.

Install time improvements

On Windows x64 machines with GPUs, we will automatically detect your GPU vendor and download only the appropriate GPU support files for your hardware. This will greatly improve the install time of AnythingLLM on Windows machines with GPUs.

This dramatically decreases the install time of AnythingLLM on Windows machines with AMD GPUs since the supporting binaries are much smaller.

Other Improvements

• Agents now report metrics and citations from docs + websites used
• OpenRouter stream metrics are now accurate
• Native tool calling for Novita
• Removed Google SERP as a default web-search provider (Google killed it)
• Add long-timeout fetch to Ollama embedder
• Better addtoworkspace errors in API
• Remove use_mlock from Ollama for API compatibility
• Added document count indicators to workspace document picker
• Perplexity search SERP is now available as a web-search provider

Bug Fixes

• Show LMStudio error state in model picker
• More Confluence Wiki fixes
• Strip thinking outputs from copy message in chat window
• Meeting Assistant - auto-verify model files before transcription to prevent errors during transcription

Pinned Download Links

Revision 1.11.2: