Locals
Run many AI agents at once on your Mac, up to 3x faster.
- agents
- apple-silicon
- artificial-intelligence
- coding-agents
- inference-acceleration
- inference-engine
- macos
- mlx-lm
- Stars
- 288
- Forks
- 17
- + today
- +1
- Created
- 18d
Ranking data as of October 29, 2026 (UTC).
Star History
Today, hour by hour
Overview
Run many AI agents at once on your Mac, up to 3x faster. It ranks #4634 on GitTiger, gaining +1 star on October 29, 2026 (UTC).
The project is written in Python and has 17 forks. It was created 18d ago.
What it is
Locals is a Python project that runs AI models on your own Mac. Its description promises running many agents at once, up to three times faster. Topics point to Apple silicon, MLX-LM, macOS, coding agents and inference acceleration. The listed license is MIT.
Capabilities described
The README summary frames it as a local server built for agents rather than plain chat. It argues that agents make tool calls that must be correct, send long prompts that grow each turn, and often work in parallel, such as coding assistants running helpers or teams of agents splitting a task. Locals is presented as designed for that workload. The three-times speedup is a claim from the description; the supplied data does not include benchmarks or conditions. The summary is truncated, so further features are unknown.
Getting started
The supplied data contains no install or setup steps, so consult the repository README for instructions.
git clone https://github.com/Generated-Computers/Locals.git
cd Locals
# see README for setupFrequently Asked Questions
What hardware does Locals target?
The description and topics point to Macs, specifically Apple silicon and macOS.
Is it a chat app?
No. The README summary says it is a local server for agents rather than plain chat.
Is the 3x speedup verified?
It is a claim in the repository description; no benchmark details were supplied.
What license is it under?
The supplied metadata lists MIT.