Back to directory

Neo Chat

A local-first, self-hostable AI chat workspace that brings multi-model chat, an agent mode for multi-step tool workflows, and deep research with reviewable plans into one app, keeping conversation history in the browser by default.

Neo Chat

Neo Chat is a local-first AI chat workspace that connects to a user's own model providers, including Google, OpenAI, Anthropic and OpenAI-compatible endpoints, with image input and generation where supported. Beyond everyday conversation, it offers an agent mode for multi-step tool workflows and a deep research mode with reviewable plans, cited reports and manual resume.

It works as a workspace for a user's own knowledge, with file attachments, knowledge-base retrieval, local search, memory and reusable assistant presets, and can be extended with text skills, OpenAPI plugins, remote MCP servers, and an optional Docker bridge for local MCP tools. It also supports voice, editable artifacts, Markdown, math, diagrams and interactive charts with export controls. Data is stored in the browser by default, with ZIP backup and restore and optional end-to-end encrypted sync through WebDAV or S3/MinIO.

It runs with Node.js and pnpm, or as a Docker container, and supports English, Simplified Chinese and Japanese.

More in Agent frameworks

See all 902
  • Recurse

    recurse.run

    A serverless harness that lets a coding agent design, test, and deploy its own specialist sub-agents as tools, MCP servers, or bots, defined by a schema-validated manifest.

  • ProofForge

    shulgin.is-a.dev

    An AI-agent pipeline that decomposes a mathematics problem, proves each piece, and formalizes the result in Lean 4 and Mathlib until the Lean kernel accepts it, with six of its proofs already merged into Google DeepMind's formal-conjectures repository.

  • local-coder

    github.com

    Builds a local coding-agent team on top of Ollama and OpenCode by detecting the machine's hardware, recommending models for seven specialist roles, and testing whether each assigned model can actually use tools, edit a repository and pass verification.