anything-llm Releases

Enhanced API capabilities, added Drupal wiki connector, and improved accessibility features.

May 6, 2025

Major UI refresh and full Model Context Protocol (MCP) compatibility.

Apr 18, 2025

Introduced system prompt variables, NVIDIA NIM support, and improved slash command functionality.

Mar 27, 2025

This release focuses on UI refinements for agent interactions and thought containers, alongside improvements to embedding endpoints (OpenAI compatible) and ChromaDB. It introduces Danish translation and the ability to reorder workspaces.

Mar 19, 2025

Major provider updates include fixing the Anthropic `/models` endpoint and adding support for the PPIO provider. Includes bug fixes for the collector, multi-language OCR support, and various localization improvements.

Mar 4, 2025

This release introduces Open-Sora V1.3, focusing on architectural improvements in video generation. Key updates include upgraded video compression, enhanced ST-DiT architecture, and improved generation quality through multi-stage training.

Feb 21, 2025

A significant feature release introducing the no-code agent skill builder and local OCR scanning for PDFs and images. The release also synchronizes Docker and Desktop versions.

Feb 18, 2025

Focuses on supporting 'thinking' models like DeepSeek and o1. Includes UI improvements for agent calling, new vector search API endpoints, and support for NVIDIA NIM and Google Generative AI.

Jan 29, 2025

Launches the AnythingLLM Community Hub for sharing system prompts and agent skills. Includes a major UI overhaul with Dark and Light mode support and new integrations.

Nov 26, 2024

Focuses on API enhancements including LiteLLM and Apipie support. Introduces new models like OpenAI o1 and Grok/XAI, along with daily message limits and improved data connectors.

Oct 25, 2024

Highlights custom agent skills, XLSX file support, and Confluence URL improvements. Includes Tavily web search and DeepSeek support.

Oct 3, 2024

The 'Custom Agent Skill Release' introducing custom agent skills and the AnythingLLM Browser Extension. Adds Elestio as a one-click deploy option and new Voyage AI models.

Sep 11, 2024

Focuses on bug fixes for slash commands, Go filetype support, and agent context window refactoring. Includes Hebrew language support and JWT encryption for single-user mode.

Aug 19, 2024

Introduces PiperTTS for in-browser text-to-speech with 100+ voices. Adds full Portuguese and German translations, MathJax/LaTeX support, and multiple slash command expansion.

Aug 8, 2024

A patch release focusing on stability and new integrations. It adds support for Gemini Pro 1.5 (including 2M context), introduces Italian translation, and improves streaming reliability for OpenRouter.

Aug 2, 2024

This release introduces multimodality support, allowing users to upload and embed documents directly within the chat window. It also includes improvements to the workspace API and various UI/UX bug fixes.

Aug 1, 2024

This is the inaugural formal release of AnythingLLM, establishing the project's versioning cycle. It establishes the core foundation including vector database support, Docker deployment, multi-user capabilities, and support for multiple LLM providers.

Jul 26, 2024

A maintenance release focusing on bug fixes for the shell and build system, alongside the addition of a system tray widget and enhanced label truncation.

Apr 21, 2024

A significant update introducing Vanilla CSS support, a new stack widget, and various mathematical and JSON processing functions.

Feb 17, 2024

A release focused on calendar widget improvements, CSS enhancements, and new input features.

Sep 4, 2022