Introducing 🏝️ OASIS: Open Agent Social Interaction Simulations with One Million Agents 🏝️ OASIS is a scalable, open-source social media simulator that integrates large language models with rule-based agents to realistically mimic the behaviour of up to one million users on platforms like Twitter and Reddit. Here are 4 of its key features: - 📈 Scalability: OASIS supports simulations of up to one million agents, enabling studies of social media dynamics at a scale comparable to real-world platforms. - 🔄 Dynamic Environments: Adapts to real-time changes in social networks and content, mirroring the fluid dynamics of platforms like Twitter and Reddit for authentic simulation experiences. - 👍 Diverse Action Spaces: Agents can perform 21 actions, such as following, commenting, and reposting, allowing for rich, multi-faceted interactions. - 🔥 Integrated Recommendation Systems: Features interest-based and hot-score-based recommendation algorithms, simulating how users discover content and interact within social media platforms. 💻 Check out the repository: ✍️ Read the paper: 🌐 Find out more via the project page: 🐫 Join our community:

🚀 Exciting News! 🚀 CAMEL AI has officially integrated with @SiliconFlowAI , a cutting-edge AI Infra platform designed for scalable, standardized, and high-performance AI model services! 🔥 With this integration, users can now seamlessly access @deepseek_ai R1 via SiliconFlow, enabling fast and cost-efficient model fine-tuning and deployment. Whether you're a developer or an enterprise, this one-stop platform empowers you to focus on innovation while we handle the AI infra for you. Try it out today and supercharge your AI workflows! ⚡ 👉 Explore more here:

CAMEL-AI🐫 Project Meeting (US Time Friendly) - Next Monday! Join us for our next development meeting to discuss our project’s latest integrations and upcoming features! 🤩🤩🤩 🚀 Exciting New Features Coming to CAMEL-AI! 🚀 We are thrilled to announce some fantastic new developments that our team will be focusing on in the upcoming weeks. Here’s a quick preview of the features we’ll be working on: - 📊Example Dataset Creation for Huggingface: We’re creating example datasets, like “math” and “CoT” dataset, and publishing it on @huggingface. - 🔍Textual Knowledge & Data Tools Integration: A new feature to improve integration with textual knowledge and data tools, enhancing our platform’s ability to work with different knowledge sources. - 👨‍💻Input Field Use Cases for @AlpacaNetworkAI : We will consider different use cases for the input field in Alpaca items, optimizing how users interact with this feature. - 🤖Map Reduce Agent Society: A new feature aimed at improving the interaction between different agents in our system. - 🌐Azure Embeddings Support: Expanding our platform’s capabilities by supporting @Azure embeddings to offer more flexibility and functionality. - 🧬Rejection Sampling Data Generation Pipeline: A data generation pipeline for rejection sampling data generation, perfect for boosting data quality and system performance. - 📩Channel Abstraction for Message Exchanges: We’re adding channel abstraction to streamline message exchanges between agents, improving communication and efficiency. - 🍳Cookbook for Benchmark Running & Source2Synth: We’re releasing cookbooks to help users run benchmarks and implement the source2synth process more effectively. - 🌌SkyThought Data Generation Method: We will integrate the information into @NovaSkyAI SkyThought distillation method to enhance data generation capabilities. - ✅Verifier Module for Evaluating Agentic Data: A new verifier module will be added to evaluate synthetic data, enhancing reliability and trustworthiness. - 💻R Language Support in Interpreters: We’re adding support for R language in our interpreters, expanding our language support for data processing. - 🏃‍♂️Batch API for Models: A Batch API will be implemented for models to facilitate more efficient processing. - 🏆Berkeley Function Calling Leaderboard (BFCL): We’re introducing the BFCL to track and evaluate function calls, providing new insights and comparisons. - 💡Reasoning Effort Parameter for @OpenAI Models: This feature will allow more control over the reasoning efforts of OpenAI models, particularly for models like o3-mini. ⏰ When: 17:00 (BST) / 09:00 (PT) 📍Where:

Learn about the "The DeepSeek Moment" in this talk! 🚀 @hallerite presented "The DeepSeek Moment", where he went in-depth on how 🐳@deepseek_ai R1 works on a technical level and how it is changing the game for all builders in AI. 👉 Watch the full talk here: 🔗 Join us live for our next event:

🤝 Join us in 2 hours for our next 🐫 CAMEL-AI Community Catch-up Come present what you've been working on this week, share ideas and meet other people interested in multi-agent systems. ⏰ When: today at 17:00 (BST) / 09:00 (PT) 🗺️ Where:

Build with DeepSeek & CAMEL-AI to win prizes! 🏆 Massive thanks to @lablabai for organising the "Fall in Love with DeepSeek" hackathon!🤝 We recommend everyone to look at our docs for some example use cases! 📃 We look forward to seeing what everyone builds! 🚀 👉 See more here: 🔗 Register here:

📢 We've just integrated Gemini's new models into the 🐫 CAMEL framework! ⚡️ Flash Lite: Gemini’s fastest and most cost-efficient multimodal model with great performance for high-frequency tasks. 💻 Pro: Gemini’s best performing multimodal model with features for a wide variety of reasoning tasks. 🤔 Flash Thinking: An experimental model that’s trained to generate the “thinking process” the model goes through as part of its response. Thanks to our contributor @ttokzzzzz for this implementation. 🤝 Explore more here:

🚀 AI agents that collaborate with humans in real-time? Discover how 🐫 CAMEL agents team up with 🤝 @humanlayer_dev to pause, ask for feedback, and get human approvals mid-task—making AI workflows more adaptive and transparent. 🌟 Huge shoutout to @xiaotian_jin for this game-changing integration! 🔗 Try it now:

@GeminiApp @lmarena_ai @YouTube This looks fantastic—congrats to the team! We'll definitely have our agents test it out. 🚀

Got value from this thread? Now it’s your turn to distill high-quality reasoning data with 🐫 CAMEL-AI + 🐋 @DeepSeek R1! 🔗 Link to the cookbook : 📌 Stay Connected: Star our GitHub repos: Explore our projects: Contact us: team@gmail.com Join our community: Follow us for more useful insights and updates! 🚀

Want a full breakdown of @deepseek_ai R1? 👉 Join us in 1 hour:

📢 Join us tomorrow for our Agentic Data Research Reading Group! 📕 @hallerite will present the: "The DeepSeek Moment", where we he will go in depth on how 🐳 @deepseek_ai R1 works and how it is changing the game for all builders in AI. 👉 Join the event here: