Show HN: Running 104GB Qwen3.8-Flash-Next on 48GB Mac with at ~12 tok/s
Summary
A Show HN-style post demonstrates running Qwen3.8-Flash-Next on a 48 GB Mac by streaming weights from SSD to memory, achieving about 12 tokens per second on warm decode with 104 GB of weights. The article covers install, memory planning, download bandwidth considerations via Hugging Face, and practical limits for memory and disk space, showing how to manage expectations and verify integrity.