M5 Ultra rumored with 1.2TB/s bandwidth, beating API speeds for local LLMs
StefanoGogioso · x · 2026-08-26
Rumors suggest the 512GB M5 Ultra, shipping in October, will feature 1.2TB/s memory bandwidth, a 50% increase over M3 Ultra. By stacking 4x M5 Ultra units, users may run models like Kimi K3 and GLM 5.3 locally at speeds exceeding 100 tok/sec, outperforming API calls. Based on extrapolations from M3 Max, the M5 Ultra is expected to deliver 110 TFLOPS (dense fp16).
More from Infra
- Copper to optical: optical backplanes seen as necessary as racks move to optics — BenBajarin · 2026-10-16
- Polymarket prices 28% odds a US state enacts data center moratorium by year-end — Polymarket · 2026-10-16
- Pennsylvania residents reject $10,000-per-household offer for massive Amazon data center — Polymarket · 2026-10-16
- PyTorch now runs natively on AWS Trainium with zero code changes — PyTorch · 2026-10-16
- Chinese labs run on ~1/10 US compute; ByteDance pays 2-4x to hoover up SEA supply — robleclerc · 2026-10-16
- Zymtrace raises $12.2M and hires ML engineers to build agents that fix GPU bottlenecks — cfregly · 2026-10-16