Mark Zuckerberg announces open-sourcing Code Llama, includes superior 70B parameter model

You can request Meta's Code LLama on its official website here.

Reading time icon 2 min. read


Readers help support MSpoweruser. We may get a commission if you buy through our links. Tooltip Icon

Read our disclosure page to find out how can you help MSPoweruser sustain the editorial team Read more

Key notes

  • Meta open-sources Code Llama with 70B parameters for code generation.
  • The tech giant’s boss Mark Zuckerberg personally announced the news.
  • It aids coding in various languages, launched in August.

Meta (formerly Facebook) has announced the open-sourcing of an upgraded Code Llama, a language model specifically designed for generating and editing code. The new version boasts a significantly larger 70B parameter model

The announcement is so important that the Meta boss himself, Mark Zuckerberg, announced it personally. In a recent Facebook post while announcing the open sourcing-, he says, “The ability to code has also proven to be important for AI models to process information in other domains more rigorously and logically.” 

Launched back in August last year, Meta’s Code Llama launched in three different sizes: 7B, 13B, and 34B for different latency requirements. Built on Llama 2, the premise is simple: put word prompts, and then the AI will respond how to code what you want and give a thorough breakdown. 

Specifically, the 7B and 13B models have been trained to insert code into existing code, enabling them to assist with code completion out of the box.

“It supports many of the most popular programming languages used today, including Python, C++, Java, PHP, Typescript (Javascript), C#, Bash and more,” Meta says. 

You can request Meta’s Code LLama on its official website here and then tick on Code Llama.

Leave a Reply

Your email address will not be published. Required fields are marked *