CINXE.COM

Gorilla

<!DOCTYPE html> <html> <head> <!-- Google tag (gtag.js) --> <script async src="https://www.googletagmanager.com/gtag/js?id=G-NRZJLJCSH6"></script> <script> window.dataLayer = window.dataLayer || []; function gtag() { dataLayer.push(arguments); } gtag('js', new Date()); gtag('config', 'G-NRZJLJCSH6'); </script> <meta charset="utf-8"> <meta name="viewport" content="width=device-width, initial-scale=1.0, shrink-to-fit=no"> <title>Gorilla</title> <link rel="stylesheet" href="https://cdnjs.cloudflare.com/ajax/libs/twitter-bootstrap/4.5.0/css/bootstrap.min.css"> <link rel="stylesheet" href="https://fonts.googleapis.com/css?family=Source+Sans+Pro"> <link rel="stylesheet" href="assets/css/Highlight-Clean.css"> <link rel="stylesheet" href="assets/css/styles.css"> <link rel="stylesheet" href="assets/css/Team-Clean.css"> <link rel="stylesheet" href="assets/css/main_page.css"> <link rel="stylesheet" href="assets/css/cards.css"> <link rel="stylesheet" href="assets/css/contact.css" /> <style> .custom-cursor-gorilla { cursor: url('assets/img/gorilla-16.png'), auto !important; } .custom-cursor-mag { cursor: url('assets/img/book-16.png'), auto !important; } .custom-cursor-trophy { cursor: url('assets/img/trophy-16.png'), auto !important; } .custom-cursor-steam { cursor: url('assets/img/fire_engine-16.png'), auto !important; } </style> </head> <body> <!-- Navigation Bar --> <div class="navbar" style="position: absolute; top: 0; right: 20px; padding: 10px; z-index: 100;; font-size: 18px;"> <a href="index.html">Home</a> <a href="blog.html">Blogs</a> <a href="leaderboard.html">Leaderboard</a> <a href="apizoo/">API Zoo Index</a> </div> <div class="highlight-clean" style="padding-bottom: 10px;"> <h1 class="text-center"> 🦍 Gorilla: Large Language Model Connected with Massive APIs</h1> <div> <p></p> </div> <div class="container" style="max-width: 768px;background:white;"> <div class="row"> <div class="col-md-12"> <h4 class="text-center" style="margin: 0px;"> <a class="text-center" href="https://people.eecs.berkeley.edu/~shishirpatil/">Shishir G. Patil*</a>, <a class="text-center" href="https://tianjunz.github.io/">Tianjun Zhang*</a>, <a class="text-center" href="https://xinw.ai/">Xin Wang</a>, <a class="text-center" href="https://people.eecs.berkeley.edu/~jegonzal/">Joseph E. Gonzalez</a> <h4></h4> <h5 class="text-center">UC Berkeley</h5> <h6 class="text-center">sgp@berkeley.edu, tianjunz@berkeley.edu</h6> </div> </div> </div> <div> <p></p> </div> <div class="buttons" style="margin-bottom: 8px;"> <a class="btn btn-primary" role="button" href="blog.html" style="background-color: #0098d4;">Blogs</a> <a class="btn btn-primary" role="button" href="https://github.com/ShishirPatil/gorilla" style="background-color: #333940;">GitHub</a> <a class="btn btn-primary" role="button" href="https://huggingface.co/gorilla-llm" style="background-color: #ffb300;">HuggingFace</a> <a class="btn btn-primary" role="button" href="https://discord.gg/grXXvj9Whz" style="background-color: #6f76ed;">Discord</a> </div> <div class="buttons" style="margin-bottom: 8px;"> <a class="btn btn-primary" role="button" href="https://arxiv.org/abs/2305.15334" style="background-color: #708090">Gorilla Paper</a> <a class="btn btn-primary" role="button" href="https://arxiv.org/abs/2403.10131" style="background-color: #B87333">RAFT Paper</a> <a class="btn btn-primary" role="button" href="https://arxiv.org/abs/2404.06921" style="background-color: #8A9A5B">GoEX Paper</a> </div> <div> <p></p> </div> <div class="container spinner-container" style="background: #e5effc;"> <div> <div class="col-md-12 api-appstore"> <h3 class="spinner-small-text-style">Systems and Algorithms for Integrating LLMs with Applications, Tools, and Services</h3> </div> <div class="col-md-12"> <div class="spinner"> <h3 class="spinner-big-text-style">Gorilla Used at <div class="spinner__text"> <span class="spinner__text--top" data-texts="Microsoft, Nvidia, Tesla, OpenAI, Linkedin, Netflix, MIT, Cisco, Anthropic, Weaviate, Cohere"></span> <span class="spinner__text--bottom" data-texts="Microsoft, Nvidia, Tesla, OpenAI, Linkedin, Netflix, MIT, Cisco, Anthropic, Weaviate, Cohere"></span> </div> </h3> </div> </div> </div> </div> </div> <div class="card-container"> <div class="row"> <div class="col-md-6"> <div class="card custom-cursor-gorilla" style="background-color: #e9edc9;"> <div class="card-inner"> <div class="card-front"> <div class="card-title">Gorilla OpenFunctions</div> <p class="card-bio">Teach LLM tool use</p> <div class="card-image"><img src="assets/img/blog_post_4_gorilla_open_function_calling.png" alt="Image 1"> </div> </div> <div class="card-back"> <p> 🎁In OpenFunctions-v2, we natively train the model to support parallel functions (generate multiple functions at a time) and multiple functions (select one or more functions). Java/REST/Python APIs are also supported for the first time with extended data types📷 <ul> <li> Read More: <a href="https://gorilla.cs.berkeley.edu/blogs/7_open_functions_v2.html">Blog</a> </li> <li>How well to other function-calling models perform: <a href="https://gorilla.cs.berkeley.edu/leaderboard.html">Berkeley Function Calling Leaderboard</a></li> <li>Play with the model online: <a href="https://gorilla.cs.berkeley.edu/leaderboard.html#api-explorer">Gorilla OpenFunctions-v2 Web Demo</a></li> <li>Check out the project: <a href="https://github.com/ShishirPatil/gorilla/tree/main/openfunctions">GitHub Code</a> </li> <li>Model (6.91B) on HuggingFace 🤗: <a href="https://huggingface.co/gorilla-llm/gorilla-openfunctions-v2">gorilla-llm/gorilla-openfunctions-v2</a> </li> </ul> </p> </div> </div> </div> </div> <div class="col-md-6"> <div class="card custom-cursor-trophy" style="background-color: #dbebe9;"> <div class="card-inner"> <div class="card-front"> <div class="card-title">BFCL</div> <p class="card-bio">Benchmarking LLMs on function calling capabilities</p> <div class="card-image"><img src="assets/img/leaderboard-summary.jpg" alt="Image 1"> </div> </div> <div class="card-back"> <p> 🏆 Berkeley Function-Calling Leaderboard (BFCL) 📊 aims to provide a thorough study of the function-calling capability of different LLMs. It consists of 2k 📝 question-function-answer pairs with multiple languages (🐍 Python, ☕ Java, 🟨 JavaScript, 🌐 REST API), diverse application domains, and complex use cases (multiple and parallel function calls). We also investigate function relevance detection 🕵️‍♂️, to determine how the model will react when the provided function is not suitable to answer the user's question. <ul> <li> Read More: <a href="https://gorilla.cs.berkeley.edu/blogs/8_berkeley_function_calling_leaderboard.html">Blog</a> </li> <li>Live Leaderboard: <a href="https://gorilla.cs.berkeley.edu/leaderboard.html">Website</a> </li> <li>BFCL Evaluation Dataset: <a href="https://huggingface.co/datasets/gorilla-llm/Berkeley-Function-Calling-Leaderboard"> HuggingFace Dataset 🤗</a></li> <li>Gradio Demo: <a href="https://huggingface.co/spaces/gorilla-llm/berkeley-function-calling-leaderboard"> HuggingFace Space 🤗 </a></li> <li>Reproducibility: <a href="https://github.com/ShishirPatil/gorilla/tree/main/berkeley-function-call-leaderboard">Github Code</a></li> </ul> </p> </div> </div> </div> </div> </div> <div class="row"> <div class="col-md-6"> <div class="card custom-cursor-mag" style="background-color: antiquewhite;"> <div class="card-inner"> <div class="card-front"> <div class="card-title">RAFT</div> <p class="card-bio">Better way to do RAG</p> <div class="card-image"><img src="assets/img/blog_post_9_raft_openbook.png" alt="Image 1"> </div> </div> <div class="card-back"> <p> RAFT: Retriever-Aware FineTuning for domain-specific RAG 🚀 Drawing parallels between LLMs and students in open-book (RAG) 📔 and closed-book exams (SFT) 🧠, we present a better recipe for fine-tuning a base LLM for RAG-focused challenges. Discover how RAFT prepares LLMs to excel with a specific document set, mirroring students' prep for finals! 🎓 <ul> <li> Read More: <a href="https://gorilla.cs.berkeley.edu/blogs/9_raft.html">Blog</a> </li> <li> Read More: <a href="https://techcommunity.microsoft.com/t5/ai-ai-platform-blog/raft-a-new-way-to-teach-llms-to-be-better-at-rag/ba-p/4084674">MSFT-META Blog</a> </li> <li> Paper: <a href="https://arxiv.org/abs/2403.10131">https://arxiv.org/abs/2403.10131</a> </li> <li> Reproducibility: <a href="https://github.com/ShishirPatil/gorilla/tree/main/raft">Github Code</a> </p> </li> </ul> </div> </div> </div> </div> <div class="col-md-6"> <div class="card custom-cursor-steam" style="background-color: #fefae0;"> <div class="card-inner"> <div class="card-front"> <div class="card-title">GoEX</div> <p class="card-bio">Runtime for executing LLM-generated actions</p> <div class="card-image"><img src="assets/img/blog_post_10_intro.png" alt="Image 1"> </div> </div> <div class="card-back"> <p> Gorilla Execution Engine (GoEX) is a runtime for LLM-generated actions like code, API calls, and more. Featuring "post-facto validation" for assessing LLM actions after execution 🔍 Key to our approach is "undo" 🔄 and "damage confinement" abstractions to manage unintended actions & risks. This paves the way for fully autonomous LLM agents, enhancing interaction between apps & services with human-out-of-loop🚀 <ul> <li> Read More: <a href="https://gorilla.cs.berkeley.edu/blogs/10_gorilla_exec_engine.html">Blog</a> </li> <li> Paper: <a href="https://arxiv.org/abs/2404.06921">https://arxiv.org/abs/2404.06921</a> </li> <li> Try it out: <a href="https://goex.gorilla-llm.com/index">Web Demo</a> </li> <li> Reproducibility: <a href="https://github.com/ShishirPatil/gorilla/tree/main/goex">GitHub Code</a> </li> </ul> </p> </div> </div> </div> </div> </div> </div> <!-- Contributions --> <div class="container news-container" style="background: #e5effc;"> <div class="row"> <div class="col-md-12 text-center"> <h2>News</h2> </div> <div class="col-md-12"> <h5>🚒 <a href="https://goex.gorilla-llm.com/index">GoEx</a>: A Runtime for executing LLM generated actions like code & API calls GoEx presents “undo” and “damage confinement” abstractions for mitigating the risk of unintended actions taken in LLM-powered systems. <a href="https://gorilla.cs.berkeley.edu/blogs/10_gorilla_exec_engine.html">Release blog</a>, <a href="https://arxiv.org/abs/2404.06921">Paper</a>.</h5> <h5>🎉 <a href="https://gorilla.cs.berkeley.edu/leaderboard.html">Berkeley Function Calling Leaderboard</a>! How do models stack up for function calling? 🎯 Read more in our <a href="https://gorilla.cs.berkeley.edu/blogs/8_berkeley_function_calling_leaderboard.html">Release Blog</a>. </h5> <h5>🏆 <a href="https://gorilla.cs.berkeley.edu/blogs/7_open_functions_v2.html">Gorilla OpenFunctions v2</a> Sets new SoTA for open-source LLMs 💪 On-par with GPT-4 🙌 Supports more languages 👌</h5> <h5> 🥇 <a href="blogs/4_open_functions.html">Gorilla OpenFunctions!</a> Drop in replacement! <a href="https://colab.research.google.com/drive/16M5J2H9F8YQora_W2PDnp120slZH-Mqd?usp=sharing">Examples</a> </h5> <h5> 🚀 <a href="https://colab.research.google.com/drive/1DEBPsccVLF_aUnmD0FwPeHFrtdC0QIUP?usp=sharing">Try Gorilla in 60s!</a> No sign-ups, no installs, just colab!</h5> <h5> 🤩 With Apache 2.0 licensed LLM models, you can use Gorilla commercially without any obligations! </h5> <h5> 📣 We are excited to hear your feedback and we welcome API contributions as we build this open-source project. Join us on <a href="https://discord.gg/grXXvj9Whz">Discord</a> or feel free to email us! </h5> </div> </div> </div> <!-- Gorilla CLI and Spotlight --> <div class="container video-container"> <div class="row"> <div class="col-md-12 text-center"> <h2>Gorilla for your CLI and Spotlight Search</h2> </div> <div class="col-md-6"> <div class="col-md-12 text-center video-section-container"> <iframe src="https://www.youtube.com/embed/4EdyWkcddPc" frameborder="0" allow="accelerometer; autoplay; clipboard-write; encrypted-media; gyroscope; picture-in-picture" allowfullscreen> </iframe> </div> <p class="text-center" style="color: black;">Gorilla powered CLI<br> Get started with <code>pip install gorilla-cli</code></p> </div> <div class="col-md-6"> <div class="col-md-12 text-center video-section-container"> <iframe src="https://www.youtube.com/embed/RMgM3tPTpXI" frameborder="0" allow="accelerometer; autoplay; clipboard-write; encrypted-media; gyroscope; picture-in-picture" allowfullscreen> </iframe> </div> <p class="text-center" style="color: black;">Gorilla Powered Spotlight Search<br> <a href="https://docs.google.com/forms/d/1iuWrfQbTb8_T5s4J75HNftPF-w1Jg1Aeg-gB64X8MJg/edit">Gorilla-Spotlight Signup</a> </p> </div> </div> </div> <!-- Figure --> <div class="container" style="background: #e5effc"> <div class="row"> <div class="col-md-12 text-center"> <h2>Vision</h2> <img src="assets/img/blog_post_1_teaser.gif" style="width: 75%;margin-bottom: 8px;" alt="Gorilla LLM logo"> </div> <em class="col-md-12">Rather have the user at the center, Gorilla enables users to interact with a wide range of services through LLMs. Gorilla is an open-source, state-of-the-art LLM that invokes API calls to interact with services! </em> </div> </div> <!-- Citation --> <div class="container" style="background: #e5effc"> <div class="col-md-6 contact-us"> <h2 class="text-center">Contact Us</h2> <form class="submit-to-google-sheet" name="submit-to-google-sheet"> <input type="text" name="Name" placeholder="Your Name" required /> <input type="Email" name="Email" placeholder="Your Email" required /> <input type="Organization" name="Organization" placeholder="Your Organization" /> <textarea name="Message" rows="6" placeholder="Your Message"></textarea> <button type="submit" class="btn-secondary2">Submit</button> </form> <span id="msg"></span> </div> <div class="col-md-6"> <h2 class="text-center">Citation</h2> <pre> <code> @article{patil2023gorilla, title={Gorilla: Large Language Model Connected with Massive APIs}, author={Shishir G. Patil and Tianjun Zhang and Xin Wang and Joseph E. Gonzalez}, journal={arXiv preprint arXiv:2305.15334}, year={2023}, } </code> </pre> </div> </div> <script src="indexPage.js"></script> </body> </html>

Pages: 1 2 3 4 5 6 7 8 9 10