Locals
Locals is a local inference server for macOS on Apple silicon that runs AI models to handle many agent requests in parallel with accelerated tool calls.
Not verified yetKnown vulnerabilities
How far pickfit got
pickfit hasn’t verified Locals yet, so there are no tested install steps.
What follows comes from its README only.
Use it
Before you install: known vulnerabilities
pickfit found 2 known high or critical vulnerabilities in the packages a fresh install brings (OSV.dev, 3 Nov 2026). Weigh this before you install it.
pypi:certifi 2026.7.22certifi bundles emSign root CAs with compromised intermediate CA keys (GHSA-6q48-6c7w-9gw2, critical, fixed in 2026.10.18)pypi:filelock 4.0.12filelock ReadWriteLock: SQLite reopens the lock database by name, so a planted symlink or journal file makes the lock user's process overwrite or unlink its own files (GHSA-v4x8-3wr5-3wq8, high, fixed in 4.2.1)
You need first
- Runtimes:
python==3.12 - A GPU is required
- At least 16 GB of RAM
- Apple silicon only
What it does
pickfit hasn’t run it yet, so these come from the README only.
- Run many agent requests concurrently batched on the GPUNot tested
- Generate tool calls and JSON that always parse via GPU grammarNot tested
- Cache conversations and long system prompts across turns and restartsNot tested
- Split tasks across a team of agents each in its own project copyNot tested
Good for
- run multiple AI coding agents in parallel on a local Mac
- handle agent tool calls with structured JSON and function calling locally
An alternative to mlx-lm, Ollama and llama.cpp.