Google announces Tensor G6, bringing 4K Portrait Video and more to the Pixel 11 series

Affiliate links on Android Authority may earn us a commission. Learn more.

Google has officially launched the Pixel 11 series and the Pixel 11 Pro Fold . Like generations before them, these new Pixel 11 flagships are all powered by a Google-made custom Tensor SoC, the new Google Tensor G6 in this case. And like every year before this, Google isn’t chasing benchmark or raw spec sheet numbers, instead focusing on the holistic experiences enabled by the SoC, with a keen focus on on-device AI features.

Under the hood, the Tensor G6 addresses some of the lingering critiques of previous generations by prioritizing power efficiency. The upgraded CPU boasts up to a 20% improvement in power efficiency compared to the Tensor G5. With the new CPU, you can get up to 25% faster web browsing and 15% faster app loading times in everyday use, if Google’s testing on pre-production hardware is to be believed.

Furthermore, Google says it has specifically optimized the chip to handle demanding mobile titles like Asphalt Legends, Disney Speedstorm, and Squad Busters, which is a surprising move given gaming has never really been Tensor’s forte. This is paired with a new custom modem that is designed to deliver higher throughput and lower power draw, though Google has remained tight-lipped about the specific vendor details.

What we could figure out from our brief hands-on with the new Pixel 11 series , the Tensor G6 sports a seven-core configuration, with one prime “C1 Ultra” core clocked at 4.1GHz, four performance “C1 Pro” cores clocked at 3.4GHz, and two efficiency “C1 Pro” cores clocked at 2.65GHz. The GPU appears to be a PowerVR C-series variant with six compute units clocked at 1.3GHz.

For reference, the Tensor G5 came with one ARM Cortex-X4 (3.78GHz) cores, five ARM Cortex-A725 (3.05GHz) cores, and two ARM Cortex-A520 (2.25GHz) cores. The GPU was a PowerVR DXT-48-1536.

The real star of the Tensor G6 silicon is the upgraded Tensor Processing Unit (TPU). Google co-designed the Tensor G6’s silicon architecture directly alongside Google DeepMind’s latest AI models. Armed with 50% more compute power and higher memory bandwidth, the TPU allows the on-device Gemini Nano model to run with substantially lower latency and higher quality.

For everyday background tasks like notification summaries and voice transcripts, the G6 runs local workloads up to 3.5x faster while consuming up to 3.5x less energy. This hardware efficiency enables the phone to handle certain AI-heavy tasks, like Gboard’s new sign-to-text translations , entirely on-device.