Meta and Qualcomm team up to launch big language models on phones

Qualcomm and Meta have announced that the Llama 2 big language model will run on Qualcomm chips in phones and PCs from 2024.

Here's What We Know

Running language models on mobile processors will allow AI applications like voice assistants to run directly on devices rather than in the cloud. This could lower the cost and improve the performance of such systems.

Llama 2 is an open source model that does many of the same things as GPT-4 or PaLM-2. That said, it can be packaged in a smaller programme to run on a phone. Qualcomm will make it available on its devices that have a dedicated processor for artificial intelligence.

The companies have worked closely together on virtual reality chips in the past. Qualcomm has also demonstrated some artificial intelligence models on its processors, such as the Stable Diffusion image generator.

Go Deeper:

• Meta and Microsoft have released Llama 2, an artificial intelligence language model for commercial use.

Source: CNBC