📢 We’ve integrated CLIP embeddings into the 📷 CAMEL framework. This integration allows us to process multimodal data inputs, improving our ability to generate embeddings from both text and image inputs. 📷 github.com/camel-ai/

We've integrated the Code Interpreter into the CAMEL framework! This provides agents with the ability to execute code. A big thanks to our contributor onemquan for their work on this. 🤝

🦀 CRAB is designed to be a general-purpose benchmark framework for Multimodal Language Model (MLM) agents. We’re excited to present CRAB Benchmark-v0, built with the CRAB framework. It includes 100 tasks across two environments (Ubuntu and Android) and has been tested with four different MLMs (GPT-4o, GPT-4 Turbo, Gemini 1.5 Pro, and Claude 3 Opus) under three distinct communication settings. 🚀 This is just the beginning! 🦀 CRAB will continue as an open-source project with more exciting features, benchmarking results, and research coming soon. Watch this video to learn more about the project. 👇

Introducing 🦀 CRAB: Cross-environment Agent Benchmark for Multimodal Language Model Agents 🦀 CRAB offers a simple, end-to-end framework for building multimodal agents, managing environments, and creating benchmarks for evaluation. It features three core components: 🔀 Cross-environment support – agents can perform tasks on both 📱Android and 💻Ubuntu. 🕸️ Graph evaluator – provides detailed evaluation metrics for agents. 🤖 Task generation – automatically creates tasks by composing subtasks. By connecting multiple devices to agents, 🦀 CRAB enhances the ability to perform human-like tasks like never before. Use 🦀 CRAB to benchmark your multimodal agents! - 👨‍💻 Check out the repository: github.com/camel-ai/crab - 📝 Read the paper: arxiv.org/abs/2407.01511 - 🌐 Find out more via the project page: crab.camel-ai.org

Introducing CRAB 🦀: watch it in action soon.

📢 We’ve added @Redisinc cache storage to the 🐫 CAMEL framework! This enhances data management with high-performance cache storage that scales to meet growing demands. A big thank you to our contributor koch3092 for this contribution. 🤝 Explore more here: github.com/camel

🎉 Celebrating 5,000 Stars on GitHub! 🎉 A heartfelt thank you to our amazing community! We couldn’t have reached this milestone without your continuous support and contributions. 🌟 What’s Next? We’re committed to leading the next generation of open-source multi-agent systems. Some of our upcoming goals include: 🤖 Enhancing agentic data generation 📷 Scaling up to 1 million agents 🚀 Expanding infrastructure to enable faster product development with CAMEL Join us on this journey: github.com/camel-ai/camel

CAMEL🐫 updates! 🛠 Tools: -Discord Bot with RAG -Redis cache -Gemini 1.5 -Docker Support -Code Interpreter -Sync to Async -Claude 3.5 Sonnet -Nemontron API 💡 Other: -OpenAI Embeddings -Docker Compose Read it all here: camel-ai.org/post/releasenot…

This extracts frames, processes them, and generates descriptive text using LLMs with multi-modal capability. Thanks to @hhhhhhuuuuiiii for this fantastic addition. 🤝 Explore more here: github.com/camel-ai/camel/pu……

📢 We’ve integrated CLIP embeddings into the 📷 CAMEL framework. This integration allows us to process multimodal data inputs, improving our ability to generate embeddings from both text and image inputs. 🔗 : github.com/camel-ai/

📢 We’ve just introduced a knowledge graph agent into the 🐫 CAMEL framework. This agent first processes the given text using @UnstructuredIO. Next, it extracts relationships between entities from the processed content, enabling more efficient data structuring.

This integration provides a consistent and isolated environment, making it easy to set up CAMEL, run examples, and develop using Docker. Thanks to @koch309 for contributing to this! 🤝 Explore more here: github.com/camel-ai/camel/bl…

📢 We’ve integrated Gemini 1.5 into the CAMEL 🐫 framework! This boosts our ability to handle long-context understanding and multi-modal data processing, simplifying large and complex tasks across text, images, and videos.

📢 We’ve just integrated @JinaAI_'s Reader API into the 🐫 CAMEL framework! This integration offers a cleaner, more efficient way to extract LLM-friendly content from URLs, significantly enhancing the quality of data fed into language models.