Skip to main content

Release 4.1 2024.7.29.1

· 3 min read

🎉 New features​

Introducing Serena​

Serena is an expert companion for creating AI Agents. It has the capabilities of generating a custom AI Agent based on user inputs in a chat, describing your application or need. Serena has the ability to decide which type of agent is best for the use cased described and creates all necesary configurations needed

Agent personality powered by AI​

Your agents personality is now simplified and powered by AI. Inside the agent designer go to the "Behaviour" tab to edit the agents personality. You can select from a variety of fields including Role, Tone of the conversation, Complexity. Upon confirmation an enhanced personality will be generated based on the atributes and additional information available on the agent. You can turn advanced mode to modify and taylor the generated personality for your specific needs.

Agent Designer - Behavior tab Configure your agent's personality form Configure your agent's personality form Configure your agent's personality form

Serenity* Star Home​

A new design of your home page provides easy access to usefull information in just one screen. From here you can:

  • Start creating a new agent with an integrated chat with Serena
  • See at a glance your most recently edited agents
  • Quickly create an agent from a template
  • Access useful resources and complementary information with direct links

Unified agent view​

Access all of your agents within a new view. This new view provides access to easy filtering for each type of agents, a search bar and a sorting filter by name, recently edited, newest and oldest.

🛠 Enhancements​

  • Agent designer system description can now be used with up to 20000 characters
  • When creating a new agent, Chat GPT 4o Mini is set by default so that you can quickly set up an start using a fast, inexpensive model out of the box without any additional configuration.

🤖 New Models​

Mistral Large 2​

Mistral Large 2 has a 128k context window and supports dozens of languages including French, German, Spanish, Italian, Portuguese, Arabic, Hindi, Russian, Chinese, Japanese, and Korean, along with 80+ coding languages including Python, Java, C, C++, JavaScript, and Bash.

Mistral NeMo​

A state-of-the-art 12B model with 128k context length, built in collaboration by Mistran and NVIDIA. Mistral NeMo offers a large context window of up to 128k tokens. Its reasoning, world knowledge, and coding accuracy are state-of-the-art in its size category.

Mistral Mamba​

Mamba2 language model specialised in code generation. Mamba models offer the advantage of linear time inference and the theoretical ability to model sequences of infinite length. It allows users to engage with the model extensively with quick responses, irrespective of the input length. This efficiency is especially relevant for code productivity use cases

🐛 Bug fixes​

  • Improved handling of knowledge files with string chains in the format {{}}
  • Added a default background image on assistant demo page if none is provided
  • Added display icon and corrected localizations on knowledge editor.
  • Fixed bug in cost manager dashboard that didn't reflected monthly percentage change
  • When Requires user identifier is activated on agent designer execution on preview uses the current user avoiding validation error.