April 24, 2024, 10:15 a.m. | code_your_own_AI

code_your_own_AI www.youtube.com

Microsoft released a new LLM, especially for Phone. The new PHI-3 MINI 3.8B. Microsoft claims a better performance than Llama 3 8B Instruct, given its new, innovative datasets.

We test PHI-3 MINI 3.8B for the first time on 6 different tasks. New Microsoft's PHI-3 models are really OPEN SOURCE (MIT License) , you do not have to accept the META LLAMA License! Nice Microsoft.

There is also a microsoft/Phi-3-mini-128k-instruct model with a context length of 128000 tokens.
https://huggingface.co/microsoft/Phi-3-mini-128k-instruct
This model …

datasets llama llama 3 llm microsoft mit open source performance phi phi-3 phone tasks test

Data Engineer

@ Lemon.io | Remote: Europe, LATAM, Canada, UK, Asia, Oceania

Artificial Intelligence – Bioinformatic Expert

@ University of Texas Medical Branch | Galveston, TX

Lead Developer (AI)

@ Cere Network | San Francisco, US

Research Engineer

@ Allora Labs | Remote

Ecosystem Manager

@ Allora Labs | Remote

Founding AI Engineer, Agents

@ Occam AI | New York