Nvidia's Vera CPU Gains Traction in AI Industry
Nvidia's Vera CPU, a general-purpose chip, is gaining adoption in the AI industry, with Perplexity becoming one of the first big AI names to use it. Vera offers single-threaded performance, which is crucial for AI inference workloads. Nvidia is also working on its next-gen Rigel Arm CPU core, which promises even higher performance.
Key points
- Perplexity, an AI search firm, will use Nvidia's Vera CPU for its agent workloads, citing speed improvements of 1.5 times over traditional CPUs.
- Nvidia's Vera CPU is designed for single-threaded performance, which is critical for AI inference workloads, according to the company.
- Nvidia is also developing its next-gen Rigel Arm CPU core, which will deliver even higher performance within the same silicon footprint.
- The Vera CPU is part of Nvidia's Vera Rubin platform, an Arm-based design with dozens of custom cores built for its next generation of accelerators.
- Nvidia has already named Anthropic and OpenAI as early users of the Vera CPU.
Nvidia's Vera CPU is gaining traction in the AI industry, with Perplexity becoming one of the first big AI names to use it. Perplexity's vice-president of enterprise infrastructure, Nate Kupp, cited speed improvements of 1.5 times over traditional CPUs for agentic coding tasks. Nvidia's Vera CPU is designed for single-threaded performance, which is critical for AI inference workloads, according to the company.
Nvidia is also working on its next-gen Rigel Arm CPU core, which promises even higher performance within the same silicon footprint. The company claims that its next-gen Rigel Arm v9.2 CPU core will deliver better instruction delivery, more L2 cache, and better memory handling.
The Vera CPU is part of Nvidia's Vera Rubin platform, an Arm-based design with dozens of custom cores built for its next generation of accelerators. Nvidia has already named Anthropic and OpenAI as early users of the Vera CPU.
The adoption of Nvidia's Vera CPU by Perplexity and other AI firms highlights the growing importance of single-threaded performance in AI inference workloads. As AI systems become more complex, the need for efficient processing of single-threaded tasks is becoming increasingly critical.
Sources
The WireByte editorial team synthesises technology news from multiple primary sources, verifies the facts, and links every source. Articles are produced with AI assistance and reviewed under our editorial policy.