AI Infra Summit: Vera Rubin + DSX push tokens per megawatt, up to 35x vs GB200
NVIDIA Blog · rss · 2026-09-16
At AI Infra Summit (8,000+ attendees), NVIDIA's Ian Buck announced: Amazon Annapurna Labs co-developing NVHBM custom HBM; d-Matrix joining NVLink Fusion; Lambda validating DSX MaxLPS with 24% more token throughput in the same power budget; Vera Rubin NVL72 + Groq 3 LPX delivering up to 35x tokens/MW vs GB200 NVL72 for long-context 2T+ parameter models (2,529 output tokens/s/user on 100K-context workloads); and SemiAnalysis AgentX results showing up to 30x throughput/MW and 45x lower cost per million tokens versus GB300 NVL72. Emerald AI also demoed AI factories as dispatchable grid resources with Silicon Valley Power.
More from Infra
- To dethrone Postgres, new databases must prioritize ecosystem integrations from day one — mgill25 · 2026-11-02
- ESP32-powered 4-inch touchscreen builds a TUI dashboard for monitoring a DGX inference cluster — CATLLM · 2026-11-02
- OpenRouter crosses 10 quadrillion annualized token run rate, doubling every 10.5 weeks — deedydas · 2026-11-02
- Palmer Luckey asks: can you use datacenter RAM for things prohibited by the operator's terms of service? — avaitopiper · 2026-11-02
- Disable MTP for 20-30% faster local LLM inference on limited RAM — InternalMode8159 · 2026-11-02
- Kioxia up 216% in a year but down 68% from its June 2026 peak — zephyr_z9 · 2026-11-02