Meta has launched Llama 3, an AI model designed to outperform Llama 2 and its competitors like ChatGPT, Google Bard and more.
Also Read: Nothing is Integrating ChatGPT in Nothing Ear, Ear(a) Buds
The AI models available in 8-billion and 70-billion parameter variants. Meta’s commitment to open-source development ensures that these models will be accessible to developers worldwide.
With support from leading cloud providers and hardware platforms, including AWS, Google Cloud, and NVIDIA.
Performance benchmarks shows the superiority of Llama 3 over competing models, such as Google’s Gemini 1.5 Pro.
The pre-trained model of Llama 3 70B outperforms Gemini 1.0 Pro across multiple benchmarks, including MMLU, BIG-Bench Hard, and DROP. The 70B Instruct model surpasses Gemini 1.5 Pro in metrics.
Users across Facebook, Messenger, Instagram, and WhatsApp will benefit from AI capabilities, including faster image generation and access to web search results. The expansion of Meta AI into multiple countries, including Australia, Canada, and Nigeria.
Despite the advancements achieved with Llama 3 Meta remains focused on continuous innovation and improvement.
The company’s approach is beyond the current iteration of Llama models with plans already underway for future iterations including Llama 4 and 5.
Meta’s Llama 3 AI models available in both 8-billion and 70-billion parameter variants, are set to revolutionize the way users interact with AI-powered services.
Also Read: Insta360 X4 Review: The Ultimate 8K 360 Camera
These models integrated with Meta AI, will be accessible across various platforms, including Facebook, Messenger, Instagram, and WhatsApp, in supported countries. Meta is adopting a first approach by making the AI model an open source.
Meta has provided benchmark scores showcasing the superior performance of Llama 3 compared to competing models.
The pre-trained variant of the Llama 3 70B model has surpassed Google’s Gemini 1.0 Pro in key benchmarks such as MMLU, BIG-Bench Hard, and DROP, demonstrating its prowess in conversational AI tasks.
The instruct models of Llama 3 have also outperformed competitors like Gemini 1.5 Pro in various benchmarks.
The adoption of a decoder only transformer architecture with improvements in tokenization and attention mechanisms, ensures enhanced inference efficiency and accuracy.
Meta has leveraged its vast dataset comprising over 15 trillion tokens sourced from publicly available data, to train the latest AI Model, enabling it to deliver more relevant and contextually accurate responses.
Meta’s vision for AI is around omnipresence and performance excellence. By embedding Meta AI powered by Llama 3 across its ecosystem including smart glasses and virtual reality headsets.
Also Read: Blackmagic URSA Cine 12K is its Most Advanced Digital Cinema
/cdn.vox-cdn.com/uploads/chorus_asset/file/24987916/Mark_Zuckerberg_Meta_AI_assistant.jpg)
