In partnership with OpenAI, Microsft has been striving to implement AI capabilities across a wide range of its products and services, while also trying to introduce smaller models for specific cases. That’s when Microsoft Research launches a new AI model known as Orca. This AI model captures by launching huge language models. According to the research paper, the new Orca AI model is designed in a way that is able to overcome the gaps of smaller models by imitating the reasoning processes of huge basic models like GPT-4.
Both Orca and the models can be optimized for some specific tasks. They can be trained with the help of large language models like GPT-4. Orca is smaller in size and this means it needs fewer computing resources to run.
The research paper says that Orca is capable of imitating and learning from relatively large language models, such as GPT-4. Orca is an AI model driven by a parameter of 13 billion and is based on Vucuna. Orca is able to learn step-by-step thought processes, explanations and a variety of complicated instructions, with the help of GPT-4.
Microsoft makes use of large-scale imitation information to encourage progressive learning through Orca. Microsoft’s new model has already surpassed Vircuna by one hundred percent with zero-shot reasoning standards such as BBH (Big-Bench Hard). The new AI model, by some claims, is 42 percent faster than AGIEval’s conventional AI models.
READ ALSO: What did Microsoft announce at its Build conference?
Speaking of reasoning abilities, Orca is a relatively smaller model, but it is said to be as good as ChatGPT on standards like BBH. Additionally, Orca presents its competitive academic exams such as LSAT, GMAT, GRE and SAT. However, it is behind GPT-4.
According to the Microsoft research team, Orca can learn more advanced language models with the help of step-by-step explanations designed by man. Orca is expected to improve his capabilities and skills.
: Google IO 2023: Check out keynote highlights, latest smartphones and other product launches here