Meta recently rolled out Llama 3.1, the breakthrough in large language models. This model is claimed to be 40 times less expensive than ChatGPT, with superior performance. These models are now live on Oracle Cloud Infrastructure and promise a new direction for AI research.

Key Features of Llama 3.1

Cost-Saving Solution:
In addition, the Llama 3.1 is far less expensive compared to its competitors, making it appealing to businesses and developers with a desire to install AI functionalities without having a financial burden.
Advanced Model Variants:
The product comes in two variants, namely the 70B and 405B models, which serve different purposes, from generating content to advanced reasoning and even multilingual translations.
Extended Context Length:
Llama 3.1 supports an extended context length of 128K, allowing for a more detailed input to generate responses that are coherent and contextually appropriate.
Multilingual Support:
The model supports eight languages, and thus developers can create applications for a global audience.
Improved Functionality:
Llama 3.1 brings native function calling to enable developers to generate structured outputs that could easily interact with multiple APIs, thus streamlining the integration process.
Thorough Testing and Validation:
Meta has put Llama 3.1 through a thorough series of over 150 benchmark datasets and significantly improved its performance metrics against its precursor, Llama 2.

Since most businesses will look for effective yet affordable AI solutions, the combination of affordability with state-of-the-art capabilities by Llama 3.1 gives it an upper hand and a challenging competition to traditional models such as ChatGPT.