MediaTek Introduces Dimensity 9600 Pro Smartphone Chip

Sep 15, 2026
The 2nm Dimensity 9600 Pro supports AI models with up to 30 billion parameters and offers higher CPU, NPU, and graphics performance.

MediaTek announced in a press release the Dimensity 9600 Pro, a flagship smartphone chip built on a 2nm process. The first phones using the chip are expected to launch this quarter.

The Dimensity 9600 Pro uses a 2 plus 3 plus 3 CPU design, with cores running at up to 4.55GHz. MediaTek reports up to 17% higher single core and 15% higher multi core performance compared with the previous generation, and says the chip cuts multi core power consumption by 61%.

The platform supports on device AI models of up to 30 billion parameters. In its dual NPU design, the NPU 1090 provides 51% higher large language model prefill performance and 55% more token generation per watt, while a second NPU cuts power use for always on AI tasks by 40%.

The G2-Ultra NX GPU offers up to 27% higher peak performance, 24% lower power consumption at peak performance, and gaming at up to 185 frames per second. MediaTek also introduced the upcoming Dimensity 9600M, with the first phones using that chip also expected this quarter.

We hope you enjoyed this article

Consider subscribing to one of our newsletters like Silicon Brief or Daily AI Brief.

Also, consider following us on social media:

AI Chips & Datacenters AI Brief AI Brief (X)
Free newsletter

Silicon Brief

Weekly coverage of AI hardware developments including chips, GPUs, cloud platforms, and data center technology.

AI Token Webinar

Whitepaper

Tensordyne Napier: What If One Rack Could Do the Work of Nine?

Tensordyne

This Tensordyne whitepaper presents Napier, an inference-focused AI processor and rack-scale system based on the company’s TDN Math logarithmic number system. It examines infrastructure requirements for large mixture-of-experts and agentic models, compares major inference architecture approaches, and details the TDN AIP processor, TDN72 pod, TDN Link fabric, and Napier Ultra configuration. The paper reports simulation-based performance, cost, and accuracy-validation results, including Tensordyne’s projected comparison of one Napier rack with a nine-rack Nvidia Rubin plus Groq deployment; the chip is reported as taped out and in fabrication.

Read more

You may also like

Free, six days a week

Daily AI Brief: the AI news that matters, in your inbox.