Innovate Futures @ Benji
patreon
Smaug 72B AI Model Become The Highest LLM Benchmark In Hugging Face
🕑 Added 2024-02-08 13:00:10 +0000 UTC
New LLM AI Video : https://youtu.be/1LAeKpx9OR4
Smaug 72B AI Model become the highest benchmark score in the latest rankings by Hugging Face, a prominent platform for natural language processing (NLP) research, a new open-source large language model has taken the crown as the best in the world.
Smaug-72B-v0.1 Model Card: https://huggingface.co/abacusai/Smaug-72B-v0.1
HuggingFace Leaderboard: https://huggingface.co/spaces/HuggingFaceH4/open_llm_leaderboard
Introducing "Smaug 72B," the brainchild of Abacus AI, a startup dedicated to tackling challenging problems in the AI and machine learning field. Smaug-72B is actually a refined version of "Qwen 72B," another powerful language model developed by the talented researchers at Alibaba Group just a few months ago.
What makes today's release truly remarkable is that Smaug 72B surpasses not only GPT-3.5 and Mistral Medium, two highly advanced proprietary large language models created by OpenAI and Mistral respectively, but also its predecessor Qwen-72B, by a significant margin in various popular benchmarks. It's a testament to the impressive capabilities of this model.
Stay tuned for more groundbreaking news in the world of AI and language models.
