oMLX

oMLX

oMLX transforms your Mac into a local LLM inference server, delivering fast text, vision, OCR, embedding, and reranker models directly from the menu bar. With tiered KV cache and OpenAI/Anthropic API compatibility, developers get sub‑5‑second responses without cloud latency.

Key features

  • Menu bar launchable local LLM server
  • Supports text, vision, OCR, embedding, reranker models
  • Tiered RAM+SSD KV cache surviving restarts
  • OpenAI/Anthropic API compatibility
  • Native Swift implementation (no Electron)

Best for

Best for

  • indie developers
  • AI researchers
  • product managers

Not ideal for

  • enterprise teams needing centralized control
  • non-technical users without command line comfort
Developer Tools Artificial Intelligence Open Source Launched 2 weeks ago

Gallery

/

Frequently asked questions

oMLX downloads and caches new models automatically via its built‑in updater, ensuring you always run the latest version without manual intervention.
Yes, oMLX acts as a drop‑in replacement; simply point your OpenAI or Anthropic SDKs to the local endpoint and your key will work unchanged.
A Mac with at least 16 GB RAM, an SSD, and a recent CPU; larger models benefit from 32 GB or more for smooth performance.

Made by

Alternatives

View all

Byterover

by Lincoln
Developer Tools

ByteRover is a fully local, file-based memory layer for agents with market-best 92.2% retrieval accuracy, that supports cloud portability, a...

discli

by Rohit Kushwaha
Developer Tools

Give your AI agent a Discord account. discli lets agents send messages, react, manage threads, moderate — all from the terminal. Built with...

Related products

Developer Tools 12 votes

A sample product for verifying the DiscoverNext Slate 3 listing and product pages.

HydraDB OSS

by Ananya Pande
Developer Tools 1 vote

HydraDB is a native graph database built for AI infrastructure — memory, ontologies, and agent context. Today, we're open sourcing the core:...

Particle News

by Naomi Plasterer
Developer Tools 1 vote

It should be easier to understand what’s happening, so why does keeping up with the news feel like work? That's where Particle comes in. Par...

Developer Tools 1 vote

Claude Code just opened a PR. But do you really know what it did? By using Bench you can automatically store every session and easily find o...

Repo Prompt

by Eric Provencher
Developer Tools 1 vote

Repo Prompt helps AI models understand your codebase without wasting tokens on irrelevant code. Context Builder analyzes your project and se...

Blober.io

by Kabue Charles
Developer Tools 5.0(1 review)

Manage blobs, files and folders across different storage providers. Blober a local-first workflow engine designed to turn "data graveyards"...