Meta Launches Code Llama, Offers Faster Way to Code

Meta introduced a new AI code-writing tool, the Code Llama, which aims to make coding faster and more efficient both for developers and beginners.

Meta Code Llama

Code Llama is a large language model (LLM) that is built on the top of Meta's next generation open source LLM, the Llama 2. The new LLM is a code-specialized version of the existing Llama 2.

"We believe an open approach to AI is best for developing new AI tools that are innovative, safe and responsible, so we're releasing Code Llama for both research and commercial use under the same community license as Llama 2," Meta wrote on a blog post.

What makes Code Llama special is its ability to generate code both from code and natural language prompts. In addition, it can be used for code completion and debugging making it easier for developers to finish their codes and fix some bugs. Meta also included two more variations such as Code Llama - Python which can take Python codes and Code Llama - Instruct which can generate helpful answers in natural language.

Meta Code Llama Availability

The Code Llama was released in three sizes with 7B, 13B, 34B parameters respectively. The 7B and 13B models are advisable to use for low latency tasks and real-time code completion. Meanwhile, the 34B model is suited for getting best results and better coding assistance.

"The goal is to make developer workflows more efficient so that they can focus on the most human-centric aspects of their job, rather than repetitive tasks," Meta explained the logic behind Code Llama.

Code Llama is available for free both for research and commercial use.