Meta AI has unveiled its latest innovation in artificial intelligence (AI), Code Llama. Meta’s latest large language model (LLM) is poised to change how code is written, understood, and debugged.
Today we’re releasing Code Llama, a large language model built on top of Llama 2, fine-tuned for coding & state-of-the-art for publicly available coding tools.
Keeping with our open approach, Code Llama is publicly-available now for both research & commercial use.
— Meta AI (@MetaAI) August 24, 2023
With its specialized versions and promising benchmark performance, Code Llama offers a potential transformation in coding practices across various sectors.
What Is Code Llama?
Code Llama builds on the well-established framework of Llama 2 and offers three distinct models:
- The foundational code model.
- Codel Llama, a version explicitly fine-tuned for Python programming.
- Code Llama – Instruct, designed to handle natural language instructions more effectively.
The model stands out for its coding capabilities and is trained to utilize popular programming languages like Python, C++, Java, PHP, and more.
The LLM is available in three different sizes with 7B, 13B, and 34B parameters, catering to varying requirements concerning serving and latency.
CodeLlama — a version of Llama2 that was fine-tuned for code tasks is live now. Available in 7B, 13B and 34B.https://t.co/4PQBlchlrn pic.twitter.com/LEz9z44IkI
— Soumith Chintala (@soumithchintala) August 24, 2023
While the 7B model is ideal for single GPU serving, the 34B model offers superior results and is more suitable for extensive coding assistance.
Risks And Responsibilities
In benchmark testing, Code Llama has demonstrated remarkable performance. Scoring 53.7% on HumanEval and 56.2% on Mostly Basic Python Programming (MBPP), it surpassed other open-source code-specific LLMs and matched the performance of models like ChatGPT.
While the technology could be valuable to developers across all experience levels and industries, Meta emphasized the importance of responsible AI usage.
The company has conducted extensive evaluations to assess the LLM’s risk of generating malicious code and provided guidelines. These guidelines align with a framework on responsible practices for synthetic media supported by Adobe, Google, Meta, Microsoft, OpenAI, and others in big tech.
The Future Of Generative AI For Coding
Code Llama symbolizes a significant advancement in AI-driven code generation and represents Meta’s commitment to an open approach to technology.
By making Code Llama free for commercial and research use, Meta helps more developers innovate safely in this rapidly evolving field.
Code Llama’s introduction could likely impact everyone in coding, from those entering the field to developers with decades of experience.
This new open-source LLM has the potential to streamline coding processes, aid in debugging, and enhance productivity, aligning with the growing demands of the tech industry.
Featured image: Ascannio/Shutterstock