LoginSign Up
← Back to all sparks
A

AnythingLLM

AI-ASSISTANTS
Velocity2.5

All-in-one private AI desktop and Docker app that lets you chat with any LLM and your own documents.

anythingllm.com ↗

AnythingLLM is morphing from a doc-chat tool into a local-first OS-level agent.

local-firstdesktop-appagentstool-callingon-devicemeeting-recording
◆Current state
Recent releases have layered an OS-level desktop overlay (v1.11.0), a meeting-recording Desktop Assistant pitched as a Granola/Otter replacement (v1.10.0), and frictionless 'no @agent needed' tool calling (v1.12.0). v1.12.1 polished the document-embedding pipeline with streaming progress and rolled out built-in app integrations for agents.
◆Where it's heading
The product is escaping the chat window. The arc from v1.10 → v1.12 is unmistakable: meetings, screen context, OS hotkey, then tool-calling that doesn't require a special invocation. AnythingLLM is staking out the local-first, privacy-preserving end of the agent market — owning the device rather than depending on a cloud orchestrator — and using free desktop-only features (overlay, assistant) to make that argument concrete.
◆Prediction
Next likely move is broader app-integration coverage and a sharper push on offline agent skills, alongside Mobile leaving the experimental flag. Expect more on-device model orchestration that ties the overlay, assistant, and tool-calling pipeline into one ambient surface.

◆Recent moves

  1. 28d ago

    Streamed document embedding and built-in app integrations

    Per-document embedding progress is now streamed so users can queue uploads, navigate away, and not lose work. The release also ships minimal-setup integrations for common apps that agents can call out of the box, smoothing the path from install to working agent.

    View source ↗
  2. 1mo ago

    Automatic Mode removes the @agent invocation; Intelligent Tool Selection arrives

    ⚡ SPARK

    v1.12.0 drops the requirement to type @agent for tool calls on supported providers — prompts without the prefix invoke tools automatically when relevant. Intelligent Tool Selection adds an arbitration layer over which tools fire. Combined, this collapses the chat/agent UX distinction that has framed AnythingLLM since launch.

    View source ↗
  3. 2mo ago

    AnythingLLM v1.11.2

    Chat-UI overhaul with a new prompt input, better citations, per-call agent metrics, document and web-search citation reporting during agent calls, in-line skill toggles, and a workspace-level model/provider switcher. Mostly catches the agent surface up to the polish of the chat surface introduced in v1.11.

    View source ↗
  4. 2mo ago

    Homepage redesign and native tool-calling rollout for local providers

    Reworked homepage shortens time-to-first-chat after onboarding, and native tool calling is now strongly recommended on local LLM providers that support it. Sets up the v1.12 Automatic Mode by establishing native tool calling as the default execution path.

    View source ↗
  5. 3mo ago

    AnythingLLM Desktop overlay: OS-level assistant on a hotkey

    ⚡ SPARK

    A free, desktop-specific OS-level overlay opens on a single keystroke and ingests the currently active application's context alongside chat, RAG, and agent features. This is the second move (after the v1.10 Desktop Assistant) toward ambient OS integration, not just a windowed app.

    View source ↗
  6. 3mo ago

    AnythingLLM Desktop Assistant launches as a Granola/Otter alternative

    ⚡ SPARK

    Desktop Assistant ships as a free, on-device meeting-recording and arbitrary-file summarizer powered by NVIDIA Parakeet and AnythingLLM's local orchestration, with full agent tool and MCP access. This release sets the direction the v1.11 overlay and v1.12 automatic tool calling are building on.

    View source ↗