Excited to talk tomorrow at 7:00PM on Gathertown. Would love to chat about BYOC and it's applications on Bardeen!

📢Excited to release the live Berkeley Function-Calling Leaderboard! 🔥 Also debuting openfunctions-v2 🤩 the latest open-source SoTA function-calling model on-par with GPT-4🆕Native support for Javascript, Java, REST! 🫡 Leaderboard: gorilla.cs.berkeley.edu/leaderboard.ht… Blog:

Last minute sprint for a paper deadline! Like being in grad school again, except in a really nice office

WILBUR Adaptive In-Context Learning for Robust and Accurate Web Agents In the realm of web agent research, achieving both generalization and accuracy remains a challenging problem. Due to high variance in website structure, existing approaches often fail. Moreover, existing

Websites come in many shapes and sizes. How can a generalist agent continuously learn the nuances of pages in-context? Wilbur automatically collects demonstrations and learns to rank fewshots by 𝗵𝗲𝗹𝗽𝗳𝘂𝗹𝗻𝗲𝘀𝘀. If it makes a mistake, it 𝗮𝗱𝗮𝗽𝘁𝘀 on-the-fly. (1/)

Introducing the Agent Arena by 🦍 Gorilla X LMSYS Chatbot Arena 🎯 How do different agents stack up in tasks like search, finance, RAG, and beyond? Which model is the most effective for agentic tasks? What tools do users prefer? Explore these questions and more! ✏️Blog:

wrote my first personal blog tldr: recent thoughts about the human brain and predicting the effects of experiences dhruvji.github.io/mind_transform…

Super excited to share our most recent browser agent launch!!

BardeenAgent literally destroyed OpenAI Operator and Claude Computer Use in web scraping benchmarks 🔥🔥 Key stats: - Recall: 66% (double the state-of-the-art) - Cost: 3x cheaper per extracted row Here's the comparison video and method they used and research paper details 👇