Meta releases the biggest and best open-source AI model yet

Meta has launched Llama 3.1, an open-source AI model that surpasses rivals, aimed to outdo ChatGPT by year's end.

: Meta introduces Llama 3.1, a powerful open-source AI model with 405 billion parameters, outperforming competitors like GPT-4o on several benchmarks. CEO Mark Zuckerberg predicts this model will make Meta AI the most widely used assistant by year's end. Llama 3.1 is available in more countries and languages, including a unique feature to generate images based on someone's likeness.

Meta has unveiled Llama 3.1, claiming it to be the largest and most advanced open-source AI model to date with 405 billion parameters and trained on over 16,000 Nvidia H100 GPUs. CEO Mark Zuckerberg anticipates that the model’s performance will lead Meta AI to become the most widely used assistant by the end of the year, potentially surpassing ChatGPT’s usage numbers.

Llama 3.1 outperforms competitors like OpenAI's GPT-4o and Anthropic’s Claude 3.5 Sonnet on various benchmarks. Meta plans to roll out this advanced model in more countries, support additional languages, and offer image generation features based on user likeness while working with companies like Microsoft, Amazon, Google, and Nvidia for deployment.

Meta remains optimistic about the potential of open-source AI, drawing comparisons to the success of open-source operating systems like Linux. The company has tested the model for potential cybersecurity and biochemical applications, further advancing its AI capabilities. The assistant will be integrated into WhatsApp, Instagram, Facebook, and the Quest headset, enhancing user interactions across Meta's platforms.

LlamaMetaAI