Meta Claims Llama 3 AI Surpasses Gemini & Claude: Exclusive on WhatsApp, Instagram, and Facebook
In a groundbreaking move, Meta has announced a significant upgrade to its open-source generative AI assistant, Meta AI, with the introduction of the more robust Llama 3 large language model. This new iteration promises to elevate the quality of AI-generated content to unprecedented levels, marking a significant milestone in the rapidly evolving AI landscape.
Llama 3 is available in two distinct versions: 8B and 70B parameters, with a higher parameter count typically translating to improved output quality. Meta’s ambitious goal is to create an open-source model that can rival the proprietary giants in the AI industry, fostering collaboration and innovation within the community.
Explore 3600+ latest AI tools at AI Toolhouse 🚀
By keeping these models open source, Meta extends an invitation to the global AI community, encouraging collaboration and feedback as these models continue to evolve and adapt to the ever-changing demands of the digital age.
To demonstrate the prowess of Llama 3, Meta subjected the model to rigorous benchmark evaluations, and the results were nothing short of impressive. In these tests, Llama 3 8B outperformed other open-source AIs, such as Mistral 7B and Gemma 7B, while the Llama 3 70B flexed its muscles against formidable competitors like the new Google Gemini Pro 1.5 and Anthropic’s Claude 3 Sonnet. Moreover, in Meta’s own human evaluation tests, Llama 3 70B emerged victorious against Claude Sonnet, Mistral Medium, and even OpenAI’s GPT 3.5, solidifying its position as a force to be reckoned with.
As Llama 3 gears up for deployment, Meta has taken a proactive approach to responsible development and deployment, adopting a “system-level approach” that combines human oversight and automation to weed out problematic responses and fine-tune the model to avoid undesirable outputs.
With an eye on the future, Meta is already setting its sights on even more ambitious models with over 400 billion parameters. As the training process continues, users can expect to witness new capabilities, such as multimodality, multilingual conversations, extended context windows, and overall stronger performance, further solidifying Meta’s position as a leader in the AI revolution.
Meta AI is already available in English across the United States on WhatsApp, and the company is actively expanding its reach to more than a dozen countries, including Australia, Canada, Ghana, Jamaica, Malawi, New Zealand, Nigeria, Pakistan, Singapore, South Africa, Uganda, Zambia, and Zimbabwe. Additionally, select users in India have also had the opportunity to experience Meta AI on WhatsApp, although access remains limited for now.
As Meta continues to push the boundaries of AI capabilities with Llama 3, the world eagerly awaits the transformative impact this groundbreaking technology will have on industries and societies alike, ushering in a new era of innovation and progress.
Don’t forget to follow us on LinkedIn. Do join our active AI community on Discord.
If you like our work, you will love our Newsletter 📰
