64GB Isn't Enough Anymore: Local AI Is Eating Through Mac RAM
gregmushen · x · 2026-10-09
@gregmushen recalls buying an M1 Max with 64GB of RAM thinking it was unlimited headroom — and now considering an upgrade, a signal of how quickly local LLM workloads are consuming memory and pushing a hardware refresh cycle.
More from Infra
- Samsung locks in 75-80% of next year's memory supply via LTAs, including HBM for NVIDIA — zephyr_z9 · 2026-10-09
- Per-token cost of Claude Pro's Opus subscription works out the same as DeepSeek Flash — teortaxesTex · 2026-10-09
- vLLM on K8s: GPU Faults Can Render a Node Unusable — Inference Is Stateful — tianyin_xu · 2026-10-09
- ComfyUI now has an official desktop app — is it worth switching from running it directly? — pleok · 2026-10-09
- Uber details its MCP Gateway, the unified platform all its AI agents use to reach backends — AxSaucedo · 2026-10-09
- 27B model on a single RTX 4090: 262K context at ~130 tok/s with NInfer — Distinct-Pie2389 · 2026-10-09