Microsoft's Maia 200 hits record 1.4 TB/s Ethernet bandwidth per chip, beating Blackwell by 1.5x
thoefler · x · 2026-10-05
Microsoft's next-gen custom AI chip, Maia 200, sets a record with 1.4 TB/s full-duplex Ethernet bandwidth per chip — over 1.5× that of Nvidia's Blackwell.
The poster notes that "bidirectional" was doing heavy lifting in NLink's advertised 1.8 TB/s figure; real unidirectional bandwidth is far lower than it appears. Inter-chip networking is becoming as critical as raw compute for large-scale training and inference.
More from Infra
- Inference Economics: Google Now Processes 3.2 Quadrillion Tokens Monthly, 7x a Year Ago — mikeflache · 2026-10-05
- InP choke pushes industry toward 1um lasers on GaAs and hollow-core fiber — jwt0625 · 2026-10-05
- Cloudflare's new Web Search API is just a wrapper around Exa and a few other providers — gaganghotra_ · 2026-10-05
- Top 10 providers on OpenRouter ranked by monthly token volume — stuffyokodraws · 2026-10-05
- Crusoe CEO explains its layered compute business: 5-year leases to high-margin inference — AccBalanced · 2026-10-05
- AMD MI355X beats Nvidia on inference margins in SemiAnalysis InferenceX benchmark — AccBalanced · 2026-10-05