Meta's Llama 3 language model outperforms competitors in tests
The Verge — April 18, 2024, 09:00 PM UTC
Summary: Meta's new Llama 3 language model, with 8B and 70B parameters, outperforms similar models like Google's Gemma and Gemini. In benchmark tests, Llama 3 excels in diversity, reasoning, and code writing. Human evaluators also rated Llama 3 higher than OpenAI's GPT-3.5. Meta plans to expand Llama 3's capabilities to include multimodal responses. Larger models with over 400B parameters are in training and show promising performance.
Article metrics
The article metrics are deprecated.
I'm replacing the original 8-factor scoring system with a new and improved one. It doesn't use the original factors and gives much better significance scores.