News

Meta Unveils AI Model for Code Writing

Published

8 months ago

August 25, 2023

Meta Platforms announced on August 24 that it is preparing to introduce a groundbreaking artificial intelligence (AI) model tailored to aid in the composition of computer code. This move underlines the company’s ongoing commitment to establishing a foothold in the realm of cutting-edge technology, as reported by Reuters.

Also Read: SABC’s Samora Mangesi Arrested for Suspected Drunk Driving

The AI model, named Code Llama, is poised to become available for users without any associated costs. It is designed to fulfil a multifaceted role:

Generating code based on human-generated text cues
Facilitating code auto-completion
Assisting in identifying and resolving errors in programming logic

This comprehensive functionality was elucidated in a recent blog post released by the prominent social media corporation.

In the wake of OpenAI’s successful launch of ChatGPT in the previous year, there has been a notable surge in the adoption of generative AI within the corporate sector. This momentum is driven by a quest for novel capabilities and an innate desire to streamline and optimise business operations.

Meta Platforms, the parent company of Facebook and Instagram, has emerged as a prolific purveyor of AI models throughout this year. Most of these models, characterised by their availability at no cost, are emblematic of Meta’s strategic approach. A prominent example is the open-source language model Llama, which has garnered considerable attention due to its potential to disrupt the dominance of proprietary counterparts championed by Microsoft-supported OpenAI and Google, a subsidiary of Alphabet Inc.

The forthcoming AI innovation, a direct offspring of Meta’s advanced Llama 2 language model, is slated to offer diverse configurations, catering to user preferences and requirements. This strategic versatility aligns with the organisation’s aspirations to compete assertively against Microsoft’s GitHub Copilot, a tool dedicated to automated code generation.

Code Llama has been meticulously crafted to support a variety of widely used coding languages, including but not limited to Python, Java, and C++. However, it is essential to note that the AI model’s proficiency is squarely confined to coding-related tasks. Meta has explicitly cautioned against its utilisation for general text-based endeavours, emphasising its specialised nature.

Meta Platforms’ latest foray into AI innovation, exemplified by the impending launch of Code Llama, signifies a significant stride in the trajectory of technological advancement. The company’s strategic utilisation of its Llama 2 language model showcases its commitment to competing in the rapidly evolving landscape of AI-driven code generation and its auxiliary functionalities.