interesting if your goal is squeezing a specific oversized model onto a mac you already have. less interesting if you want a general local model runtime, but the narrow scope is at least stated plainly.
Slotstream
Run large AI models on Macs with limited RAM by streaming from SSD
9 ratings3.65
Saves240
CategoryAI
PlatformWeb
PricingFree
Slotstream enables running the 105 GB Qwen3.8-Flash-Next AI model on Apple Silicon Macs with as little as 16 GB RAM by streaming model 'experts' from SSD storage. It provides command-line tools, APIs, and a Swift library with offline use after one-time download, compatible with Ollama API. Targeted at independent developers and Mac users needing local large model inference without cloud dependency.
Ratings & Reviews
Write a reviewInformation
CategoryAI
PlatformsWeb
PricingFree
AccountNot confirmed
ListedSeptember 2026
WebVisit
Report a problem
Around the Web
Show HN240 source points · 0 comments