Running 104GB Qwen Model on 48GB Mac via SSD Streaming
carloslfu · hn · 2026-09-02
The author presents slotstream, a method to run the 125B parameter Qwen3.8-Flash-Next 4-bit model on Macs with as little as 16GB RAM—normally requiring 100GB+ memory. It leverages expert-offloading and SSD streaming. Built with Apple's MLX and Swift, it features an auto-mode for balancing memory and speed, with speculative decoding support planned.
More from Infra
- SGLang Summit sponsors grow from 11 to 36 in one week; Lip-Bu Tan, Lilian Weng to speak — BanghuaZ · 2026-10-18
- Yacine jokes his datacenter KPI is gigawatt-hours: get paged if not burning enough power — yacineMTB · 2026-10-18
- 1.68x inference speedup drops, directly accelerating GRPO post-training — yacineMTB · 2026-10-18
- OpenRouter is only a fraction of the total addressable market, poster argues — sloppenheimer · 2026-10-18
- Developer builds change alerts for LLM prices and retirements across Bedrock, Azure and Vertex — jsandhu0780 · 2026-10-18
- 12.3M tokens in 27 minutes for $0.14: agent session with 99.6% cache hit — teortaxesTex · 2026-10-18