July 22

CAMEL Release Notes [Sprint 5 & 6]

Exciting Updates from CAMEL-AI: New Integrations and Features!

Hey everyone! We're thrilled to share this week's updates, bringing in new integrations and features to enhance our framework's capabilities in multi-modal data processing, code execution, and more. Here's a quick rundown of the latest additions:

πŸ›  Tool updates:

  • πŸ›  Build a Discord Bot with RAG: A Discord bot powered by CAMEL's 🐫 agent and RAG pipeline is now available, providing responses based on user knowledge bases in Discord channels. Thanks to willshang76 for this improvement.  🀝 Explore more here.

‍‍‍

  • πŸ›  Redis cache storage: We've integrated Redis cache storage, enhancing data management and persistence with high-performance, scalable technology. Thanks to koch3092 for this improvement. 🀝 Explore more here.
  • πŸ›  Gemini 1.5: We’ve integrated Gemini 1.5 into the CAMEL 🐫 framework, boosting our long-context understanding and multi-modal data processing for text, images, and videos. Big thanks to Asher-hss for this significant enhancement. 🀝 Explore more here.
  • πŸ›  Add Docker Support for Code Execution: We've enabled code execution in Docker, ensuring isolated and secure environments for running scripts in multiple languages. Thanks to WHALEEYE for this update. 🀝 Explore more here.
  • πŸ›  Code Interpreter: Code Interpreter is now a tool within our framework, enabling dynamic code execution for agents. Thanks to onemquan for this feature. 🀝 Explore more here.
  • πŸ›  Sync to Async Conversion Utility: The new sync_funcs_to_async utility converts synchronous functions to asynchronous, ensuring smooth, concurrent operations. Thanks to zechengz for making this possible. 🀝 Explore more here.
  • πŸ›  Claude 3.5 Sonnet: We’ve integrated Anthropic AI's Claude 3.5 Sonnet model, excelling in reasoning, coding, and visual tasks. Thanks to Wendong-Fan for this fantastic update. 🀝 Explore more here.
  • πŸ›  Nemontron API Integration: We've integrated Nemotron-4 340B Reward Model from Nvidia, Nemotron-4 340B Reward Model is a state-of-the-art multidimensional Reward Model. The model takes a text prompt as input – and returns a list of floating point numbers that are associated with the five attributes in the HelpSteer2 dataset, Nemotron-4 340B Reward can align with human preferences for a given prompt and is therefore able to replace a large amount of human annotations. Thanks to Wendong-Fan for this implementation. 🀝 Explore more here.

πŸ’‘ Other updates:

  • πŸ’‘ OpenAI Text Embeddings: We’ve updated text embedding functionality to align with OpenAI's latest models, enhancing capabilities with text-embedding-3. Thanks to zechengzh for this great work. 🀝 Explore more here.
  • πŸ’‘ Docker Compose Support: Docker support is now available for installing the CAMEL 🐫 framework, providing a consistent and isolated environment for easy setup and development. Thanks to koch3092 for this contribution. 🀝 Explore more 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.

πŸ“˜ 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.

πŸ™Œ 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!