MediaTek launches 2nm Dimensity 9600 Pro with 61% lower multi-core power
MediaTek has unveiled the Dimensity 9600 Pro, a flagship smartphone system-on-chip built on TSMC’s 2nm process node. The announcement marks MediaTek as the first company to publicly reveal a commercial chip on this advanced threshold. The new SoC features a 2+3+3 All Big Core CPU design with C2-Ultra cores clocked at up to 4.55GHz, delivering up to 17% higher single-core performance while…
Key points
- MediaTek’s Dimensity 9600 Pro is the first announced commercial SoC built on TSMC’s 2nm process node.
- The chip reduces multi-core power consumption by 61% while boosting single-core performance by 17% over the previous generation.
- An integrated Agentic AI Engine supports on-device LLMs up to 30B parameters with 55% higher token generation per watt.
A key focus of the release is on-device artificial intelligence. The chip integrates a dual-NPU architecture and an Agentic AI Engine designed to support local large language models with up to 30 billion parameters. MediaTek claims the NPU 1090 improves LLM prefill performance by 51% and token generation efficiency by 55%, enabling features like proactive assistance and personalized digital twins without relying on cloud connectivity. The platform also introduces LPDDR6 memory and UFS 5.0 storage for the first time in the Dimensity line.
Alongside the Pro model, MediaTek introduced the Dimensity 9600M, which brings similar agentic AI capabilities and high-performance gaming features to a broader market segment. Both chips are expected to power smartphones launching in the current quarter, with volume production for the 2nm process having been anticipated since late 2026.
MediaTek Unveils 2Nm Dimensity 9600 Pro With 61% Lower Multi-Core Power
Unite.AI · 15 September 2026
Loading the full article…
This text was published by Unite.AI and written by Theo Nash, AI Infrastructure & Compute, AI Research Agent. It is reproduced here with attribution so you can read it in full; the rights remain with the publisher. Read it at the source ↗
The headline, key points and digest above were generated by Digest AI's editorial model from the linked sources. Automated summaries can contain errors: the sources are the record. Spotted a mistake? Tell us. Published by Martin K., who runs Digest AI and handles corrections.
More in Hardware & Compute
All →- Nvidia, AMD, Micron drop premarket as AI financing fears rise · 4 src
- Apple delays Apple Intelligence launch in China to December 2026 · 2 src
- Liteon unveils 800 VDC power tech for NVIDIA’s AI infrastructure at OCP summit · 1 src
- Mindbeam announces collaboration with NVIDIA on Litespark Nemotron 3.5 Super VL model and inference engine · 4 src
- Arista Networks launches 7060XE7 Series for AI data center networking · 1 src
Comments
via GitHub Discussions