Nvidia says that its next-generation AI chip platform will be launched in 2026

Nvidia’s CEO, Jensen Huang, announced at Computex that the company will release its next-generation AI chip platform, Rubin, in 2026. It will include new GPUs, CPUs called Versa, and networking chips, as well as a shift to yearly releases and partnerships for high-bandwidth memory.

Nvidia‘s CEO, Jensen Huang, announced on Sunday that the company will launch its next-generation AI chip platform, Rubin, in 2026.

Huang spoke at the Computex trade show at National Taiwan University in Taipei.

The Rubin family of chips will have new graphics processing units (GPUs), central processing units (CPUs), and networking chips.

The new CPU will be named Versa, and the new graphics chips that power AI apps will come with next-generation high-bandwidth memory from companies like Samsung, Micron, and SK Hynix.

Nvidia Accelerates AI Chip Release Cycle to Annually

Huang said that the company plans to release a new family of AI chips every year, instead of about every two years as it used to.

With about 80% of the market for AI chips, Nvidia is in a unique position as both the biggest driver of AI development and the biggest beneficiary of it.