NVIDIA details Vera CPU with 2x performance claims and a 22,000-core rack
ryanshrout · x · 2026-07-22
NVIDIA has lifted the embargo on its Vera CPU deep dive, framing the chip as a direct challenge to traditional x86 datacenter design.
- Vera uses Olympus, NVIDIA’s first custom datacenter CPU core, with a 10-wide front end and aggressive reordering/prefetching.
- NVIDIA says the design delivers roughly 2x core performance, 3x core-to-core bandwidth, and 40% lower memory latency via its LPDDR5X subsystem.
- The company describes a monolithic compute die, second-generation coherency fabric, and a rack-scale deployment option packing 256 CPUs and 22,000+ cores.
- NVIDIA’s argument is that agentic AI changes CPU demand: agents bounce between GPU and CPU for tool calls, SQL, APIs, and scripting, making CPU performance a bigger battleground.
Related event: NVIDIA Reveals More Vera CPU Details Ahead of AMD Event(8 posts)→
More from Infra
- 80% of the DIY LLM inference hype posters have already quit — it's brutally hard systems work — abhijithneil · 2026-09-11
- Direct link: Hugging Face's Ultra Scale Playbook on GPU-cluster LLM training — mdancho84 · 2026-09-11
- Hugging Face's Ultra Scale Playbook: a free book on training LLMs on GPU clusters — mdancho84 · 2026-09-11
- Is inference latency becoming the biggest bottleneck for production AI agents? — Euphoric_Sea632 · 2026-09-11
- LLM Serving Metrics Thread: Why TPOT and Uptime Make or Break User Experience — abhijithneil · 2026-09-11
- PlanetScale launches sharded Postgres: 768 servers acting as one, 1PB scale — dhruv2038 · 2026-09-11