OpenViSTAI · Agents · Blockchain · Crypto
Submit a project
Menu
Local & Self-hosted AI

Server Nexe

Server Nexe is an open-source local AI server that combines inference through MLX, llama.cpp or Ollama with persistent memory and retrieval-augmented generation. It is distributed as a Tauri desktop app and can also be installed for command-line or headless use.

Explore the practical guide ↓

Independent OpenViST project profile, based on linked sources.

Server Nexe — OpenViST project profile in Local & Self-hosted AI.
OpenViST project illustration. See the linked sources for product details.
Understand the project

Practical overview

When it fits

Use Server Nexe when you want local inference paired with persistent memory or document retrieval, or need a local chat endpoint rather than a hosted AI service.

Who it is for

  • People who want to run an AI assistant locally with persistent memory.
  • Users who want to retrieve information from documents they upload into a chat session.
  • Developers or operators who need a local OpenAI-compatible chat endpoint for applications, scripts or CI.

Documented uses

  • Run chat inference on a local model using a supported backend.
  • Store and recall facts through the memory feature.
  • Upload supported documents and use RAG to retrieve information from them within a session.
  • Provide local chat requests through the OpenAI-compatible endpoint.
Before you choose

Requirements & limits

The repository is public and licensed under Apache-2.0. Listed desktop packages cover macOS on Apple Silicon, Linux ARM64 and Windows ARM64; the repository also documents command-line and headless installation. Installation generates an API key required for API access.

  • Hardware and operating system determine which inference backends and models are available.
  • The stated minimum is 8 GB RAM and 10 GB free disk space; larger models call for more resources.
  • Uploaded documents are session-isolated rather than shared across sessions.
  • Windows is documented as Ollama-only, while MLX is limited to macOS on Apple Silicon.
  • The desktop release targets listed here are architecture-specific; check the project requirements before installing.

Check current requirements in the official documentation. Source-based guidance does not mean OpenViST has run or independently tested this project.

Source-based practical guide

Install and start Server Nexe from the command line

Set up the project with its guided installer and start the local server.

Before you start

  • A supported system and suitable inference backend.
  • At least 8 GB RAM and 10 GB free disk space; more is recommended for larger models.
  • Python 3.11 or later when installing from source.

Steps

  1. Clone the repository and change into its directory.

  2. Run ./setup.sh to detect hardware and select a backend and model.

  3. Run nexe go to start the server on port 9119.

Expected outcome

The local server starts on port 9119; its documented local endpoints include the chat API, web UI, health check and API documentation.

Guide limitations

  • Backend availability depends on the operating system and hardware.
  • The API requires an X-API-Key generated during installation and stored in .env.
Documentation 1 ↗
Evidence and transparency

Sources & review

Official repository ↗
Source record details
Source availability
Active
Sources checked
14 October 2026
Original website title
jgoy-labs/server-nexe · Official homepage title · retained in its source language
How we review sourcesSuggest a correction