Local AI for Apple Silicon

Local AI, built for your Mac.

A native studio for running compatible local models with private chat, clear setup, and a Metal-first engine that can stream from SSD.

Download for macOSView source

macOS 13+ / Apple Silicon / v0.4.0 development preview

Hebrus StudioLocal / private

Hebrus
Studio

Metal first.SSD when memory ends.

Private by default

Chats, prompts and model files stay on your Mac.

Built for Metal

Apple Silicon is the primary target, not a fallback.

SSD assisted

Large MoE models can keep moving when memory gets tight.

The Studio

Less setup. More running.

Hebrus Studio gives the engine a calm Mac surface: choose a model, start the runtime, and chat locally.

Model library

Bring your GGUFs. Know what runs.

Hebrus scans local files, checks compatibility, and keeps model setup visible.

  • 01Local discovery
  • 02Exact compatibility
  • 03Revision-pinned sources

Runtime

One switch for the local engine.

Turn it on, see the active model, and understand the current memory mode.

  • 01Metal execution
  • 02Adaptive residency
  • 03Visible runtime state

Hebrus Engine

Metal first. SSD assisted.

A focused inference path for supported large MoE models on Apple Silicon, with telemetry that says what is actually happening.

01

Metal execution

Fast local inference for the hardware in your Mac.

02

Adaptive SSD

Resident or streamed experts, selected around your machine.

03

Honest status

Endpoint, model, privacy and memory mode remain visible.

Development preview

Download Hebrus Studio.

Try the Apple Silicon preview today. The current build is ad-hoc signed while the Developer ID notarized release is being prepared.

Download development DMG

macOS 13+ / Apple Silicon / 126 MB / v0.4.0

Drag to Applications, then Control-click Open once for this preview.

Open source

Public work, credited lineage.

Hebrus began as a fork of Salvatore Sanfilippo's DS4. The project also credits llama.cpp / GGML and MLX as technical references.

Engine source Studio source