Oct 7, 2024

CAMEL Release notes [Sprint 10]

Multimodal integration, enhanced tool support, and AI scalability breakthroughs.

✨Model updates

- Integrated Reka models:Efficient, natively multimodal models trained on thousands of GPUs. From Reka Core, which rivals industry giants, to Edge for on-device use and Flash for speed, Each model is designed for specific needs and is competitive across key metrics. Thanks to our contributor Wendong-Fan for working on this. 🤝 Explore more here.

- Integrated SambaNova Systems: By doing so, we bring world-record speed and scalability to AI tasks, making real-time responses more efficient, try it out with llama3.1 models from SambaNova! Thanks to our contributor Wendong-Fan for working on this. 🤝 Explore more here.

- Integrated Together AI: Their decentralized cloud services empower developers and researchers at organizations of all sizes to train, fine-tune, and deploy generative AI models. Thanks to our contributor Wendong-Fan for this implementation. 🤝 Explore more here.

🛠️Tools updates

- Integrated the LinkedIn Toolkit: By doing so, we enhance user productivity by automating LinkedIn tasks such as posting, editing, or deleting content and fetching profile information effortlessly. Thanks to our contributor Neil Johnson for this integration. 🤝 Explore more here.

- Updated the Mistral version and added support for tool calling:By doing so, we support function calling record to enhance LLM event tracking and tool calling capabilities, ensuring more comprehensive monitoring of AI agent operations. Thanks to our contributor Wendong-Fan for this implementation. 🤝 Explore more here.

🤖️Agent updates

- Added AgentOps support for Mistral models: This enhancement provides effective evaluation, real-time monitoring, and cost management, streamlining AI agent operations. Thanks to our contributor Wendong-Fan for this valuable integration. 🤝 Explore more here.

💡Other updates

- Added Element Input Support to the Auto Retrieval Pipeline: This new feature enhances the system's ability to automatically retrieve elements (like unstructured text) as input. A huge thanks to our contributor Wendong-Fan for working on this. 🤝 You can check out the full details here.

🐫 Thanks from everyone at CAMEL-AI

Hello there, passionate AI enthusiasts! 🌟 We are 🐫 CAMEL-AI.org, a global coalition of students, researchers, and engineers dedicated to advancing the frontier of AI and fostering a harmonious relationship between agents and humans.

📘 Our Mission: To harness the potential of AI agents in crafting a brighter and more inclusive future for all. Every contribution we receive helps push the boundaries of what’s possible in the AI realm.

🙌 Join Us: If you believe in a world where AI and humanity coexist and thrive, then you’re in the right place. Your support can make a significant difference. Let’s build the AI society of tomorrow, together!