Google: The impact of running AI models on phones on memory
Latest Updates
The impact of running AI models on phones
Google announced that running artificial intelligence models on smartphones consumes a lot of random memory. The American company presents the artificial intelligence modelGemini Nano is available on its Pixel 8 Pro, but it's not available on the Pixel 8 due to hardware spec limitations..
Update details
Introducing the mod in the Pixel 8 Pro as a developer option means that users will only be able to enable it through the hidden developer options menu in Settings, rather than having it available in the UI by default. The Pixel 8 Pro has RAM(RAM) capacity is 12 GB, which makes it suitable for running the model Gemini NanoWhile the Pixel 8 has a smaller 8GB RAM..
Company Directions
Google doesn’t want to compromise the experience on the Pixel 8, which has less RAM than its predecessor. The company wants some of its AI-powered features like Smart Reply to remain “in RAM,” meaning the model will permanently take up some of the RAM..
Future prospects
Leaks indicate that Apple is working on increasing the RAM and storage capacity in the basic versions of the upcoming iPhone 16 phones in preparation for including artificial intelligence models capable of working in those devices.
Questions and Answers
1. Does running AI models affect phone performance?
Yes, studies show that running AI models consumes a lot of RAM, which may affect the performance of smartphones..
2. What steps can be taken to improve the performance of phones when running AI models?
Possible steps include including larger RAM in devices, improving software to reduce memory consumption, and improving the management of computing resources..
3. What are the trends of phone manufacturers in facing the challenges of running AI models?
Companies are moving towards increasing the RAM capacities in new devices, in addition to improving performance and software improvements to reduce the impact of memory consumption..