Scalaris Atlas

What it does

API compatibility

Atlas accepts OpenAI-compatible clients at localhost and exposes chat completions, embeddings, and model-list routes.checked 8 Oct 2001

API limits

Atlas exposes a strict subset of the OpenAI HTTP surface, and unimplemented endpoints return a typed 501 response.checked 8 Oct 2001

Configuration

Atlas configuration is stored in one TOML file, and settings reload when that file changes.checked 8 Oct 2001

Installation

Atlas is installed and supervised through Scalaris Toolbox.checked 8 Oct 2001

Integration

The maker says to point an OpenAI-compatible client at localhost:8787 to integrate Atlas.checked 8 Oct 2001

Intended users

Atlas is for people using OpenAI-compatible clients who want to run AI models locally.checked 8 Oct 2001

Local processing

The maker says models are served locally and nothing leaves the machine.checked 8 Oct 2001

Model support

Atlas includes Pulsar and lets users install their own models through its model manager.checked 8 Oct 2001

Network controls

Egress is denied by default, and Atlas refuses non-loopback binds unless the user explicitly changes the setting.checked 8 Oct 2001

Resource handling

When GPU memory is exhausted, Atlas falls back to CPU only if the user enables --allow-cpu.checked 8 Oct 2001

Telemetry

The maker says Atlas has no telemetry and does not phone home.checked 8 Oct 2001

What it does

Atlas turns a computer into a private AI host that serves local models to compatible clients.checked 8 Oct 2001