TrendingRisingGemsRadarToolboxKitsNewsCategoriesAbout

Repository

carloslfu/slotstream

Run a 105 GB AI model on a Mac that can't hold it. Slotstream streams Qwen3.8-Flash-Next (125B mixture of experts) from your SSD and caches the busiest experts in memory, so it runs on Macs with 16 to 64 GB. One native Swift binary on MLX and Metal, no Python, offline. Works with Claude Code, Codex and Ollama or OpenAI clients.

Open source AI & agents SwiftMIT

Stars

569

Today

+86

last 24 hours

This week

+89

Forks

48

Clone

git clone https://github.com/carloslfu/slotstream.git

Star growth

2026-10-22: +32026-10-23: +02026-10-24: +02026-10-25: +02026-10-26: +86 (today so far)2026-10-22todaymax +86/day
Stars gained per day since 2026-10-22; the lighter bar is today so far.
#apple-silicon#llm#llm-inference#local-llm#macos#mixture-of-experts#mlx#ollama#qwen#swift
Created
2026-08-28
Last commit
2026-10-26
Gained since tracking
+89

Similar in AI & agents

Updated Mon, 26 Oct 2026 21:02:34 GMT · description from the project itself