From 9to5google.com

Qualcomm Snapdragon 7+ Gen 3 brings on-device AI and Wi-Fi 7

Qualcomm is following its announcement at the start of this week with another chip: the Snapdragon 7+ Gen 3.

Following the 8 Gen 3 last year, Qualcomm has been working to bring premium/flagship capabilities down the range. Most notably the Snapdragon 7+ Gen 3 — the regular 7 was announced in November — allows for on-device generative AI, like images and voice assistants. Compatible LLMs include Gemini Nano, Baichuan-7B, Llama 2 and Zhipu ChatGLM. 

The 4nm Snapdragon 7+ Gen 3 features one Prime Cortex-X4 (up to 2.8 GHz), four Performance (2.6 GHz), and three Efficiency (1.9 GHz) cores. This series of chips is aimed at high-tier smartphones. 

Compared to the 7+ Gen 2, there is 15% improved Kryo CPU (Geekbench 6 Multi Thread) and 45% better Adreno GPU (GFX Aztec Ruins 1080p) performance, while being 5% more power efficient. 

In terms of connectivity, Wi-Fi 7 is coming to the 7-series with the FastConnect 7800, while the Snapdragon X63 5G Modem supports 3GPP Release 17.

On the gaming front, there’s a “Game Post Processing Accelerator” to offer “bloom, depth, and motion blur” effects, while the Adreno Frame Motion Engine 2.0 “enhances content at twice the frame rate without draining power.”

Photography with the Spectra 18-bit Triple Cognitive ISP (up to 200MP)is a big tentpole with features like semantic segmentation (for identifying layers: faces, hair, clothes, sky, etc.) and low light vision.

OnePlus will be one of the first partners to adopt the Snapdragon 7+ Gen 3, along with realme and SHARP, “in the next few months.”

Add 9to5Google to your Google News feed. 

FTC: We use income earning auto affiliate links. More.

[ For more curated Computing news, check out the main news page here]

The post Qualcomm Snapdragon 7+ Gen 3 brings on-device AI and Wi-Fi 7 first appeared on 9to5google.com

New reasons to get excited everyday.



Get the latest tech news delivered right in your mailbox

You may also like

Subscribe
Notify of
0 Comments
Inline Feedbacks
View all comments

More in computing