Microsoft has recently introduced a new small language model called Phi-3 mini, which is designed to run on modern smartphones and offer performance similar to OpenAI’s GPT-3.5. This new iteration of Microsoft’s lighter language model was trained with 3.3 billion tokens from “larger and more advanced” data sets compared to its predecessor model, Phi-2, which was trained with 1.4 billion tokens.
The updated model consists of 3.8 billion parameters, making it suitable for use in modern smartphones as it only occupies around 1.8GB of memory and can be quantified to 4 bits, according to a text published on Arxiv.org. Researchers tested Phi-3 mini on an iPhone 14 with an A16 Bionic chip and found that it runs natively and completely offline, achieving more than 12 tokens per second. The overall performance of this model “rivals” that of larger models like Mixtral 8x7B and GPT-3.5.
Phi-3 mini utilizes a transformer decoder architecture that supports a 4K text length and is based on a block structure similar to Meta’s Llama 2, benefiting the open-source community and supporting all packages developed for Llama 2. The model aligns with Microsoft’s robustness and security values by supporting a conversational chat format.
In addition to Phi-3 mini, Microsoft has also trained two other models from the same family: Phi-3 medium with 14 billion parameters and Phi-3 small with 7 billion parameters, both trained with 4.8 billion tokens. The technology company’s emphasis on innovation and performance in the field of language models is evident in their latest offerings.
Overall, the introduction of Phi-3 mini marks a significant milestone in the development of language models capable of running efficiently on mobile devices while maintaining high levels of performance comparable to larger models.
Microsoft’s commitment to innovation in the field of language models continues as they launch their latest offering – Phi-3 mini – a smaller version that runs natively on modern smartphones while delivering performance similar to larger models like GPT-3
