jundot/omlx

LLM inference server with continuous batching & SSD caching for Apple Silicon — managed from the macOS menu bar

4 best points · 1 comments · 8 Hacker News posts · 23,430 GitHub stars · View on GitHub

Hacker News discussions

• Show HN: oMLX – SSD-backed KV cache cuts coding agent TTFT from 90s to 1s on Mac (4 points, 1 comments, by jundot, 2026-03-04)
• OMLX: Local LLM Server for Apple Silicon Macs (4 points, 0 comments, by steveharing1, 2026-05-11)
• OMLX – LLM Inference Server for Apple Silicon (Ollama for MLX) (3 points, 0 comments, by fintechie, 2026-02-13)
• Show HN: oMLX – coding agents on local LLMs without the painful reprefill (3 points, 0 comments, by jundot, 2026-02-24)
• OMLX – Ollama for MLX (LLM Inference Server for Apple Silicon) (2 points, 0 comments, by fintechie, 2026-02-15)
• OMLX v0.3.9 Stable Merges Native MTP (Multi-Token Prediction) (2 points, 0 comments, by febed, 2026-05-22)
• Show HN: oMLX – Native Mac inference server that persists KV cache to SSD (1 points, 0 comments, by jundot, 2026-02-19)
• OMLX: Use a second Mac to speed up local LLM prefill 2.9x (1 points, 0 comments, by joshcsimmons, 2026-11-07)