Nvidia unveiled new AI inference accelerators and network infrastructure products at Hot Chips 2026 in the United States on Aug. 25, a day before its second-quarter earnings. The company said interactive inference accelerator Nvidia Grok 3 LPX has entered mass production, alongside Spectrum-X Multiplane Ethernet scaling and ScaleIn infrastructure acceleration technology. The announcements focused on inference, particularly token generation speed, and also highlighted inference speed, network scalability and agent-focused CPUs within the Vera Rubin platform.

Nvidia is deploying its in-house Vera CPU in semiconductor design processes to accelerate development of next-generation chips. The company said it worked with Cadence and Synopsys to optimise core electronic design automation applications. As chip design grows more complex, key EDA workloads such as logic simulation and formal verification still rely heavily on CPU performance. In tests, Cadence Jasper and Synopsys VCS delivered up to 1.5 times higher performance on some workloads with the same core count.