NEW on iPhone: SUPER AI Agents (Stanford)

New LLMs w/ functional token fine-tuning for Phones (iPhone, Pixel).
New tech increases energy efficiency of LLM function calls significantly.

Stanford Univ test new on-device LLMs w/ function calling, or tool use, for AI agent accuracy and inference speed.

Plus function calling code for OpenAI, Anthropic /Claude 3 and Cohere Command R PLUS.

00:00 On Device LLM
00:20 Technology Function Calling (Apple, Google)
15:45 CODE (Anthropic, Cohere) Function Calling
24:24 Implications for NVIDIA, Microsoft?

#airesearch
#ai
#pythonprogramming

accuracy agent agents ai agents anthropic claude claude 3 code cohere cohere command command efficiency energy energy efficiency fine-tuning function functional inference iphone llm llms openai phones pixel speed stanford tech test token tool