Product updates

Happy new year! We're kicking off 2026 with some powerful improvements and fixes that the team shipped since our last update. Stay tuned for more releases soon!

We've added two powerful new categories of tools that unlock entirely new use cases for your agents. Track your investments and monitor your brand reputation — all automated.

Check out the new Crypto Portfolio Monitor and Brand Reputation Monitor templates to get started!

Your agents just got a lot smarter. We've upgraded to Google Gemini's native function calling, which means agents now understand intent rather than just matching keywords.

This means more accurate responses, fewer irrelevant results, and agents that truly understand what you're asking for.

We've completely redesigned the layout with a focus on clarity and usability. The new design follows modern UI principles with clean typography, consistent spacing, and a refined color palette.

Don't want to configure an agent from scratch? Choose from our growing library of pre-configured templates. Each template comes with optimized tools, instructions, and schedules for specific use cases.

Just pick a template, customize if needed, and deploy. It's that simple.

Building agents just got a whole lot easier. Instead of filling out forms, you can now configure agents through natural conversation.

The AI understands your intent and updates the configuration automatically. You can see changes in real-time as you chat.

Your agents can now work together as a team. The new orchestrator intelligently routes your questions to the most relevant agents, coordinates their responses, and synthesizes everything into a unified answer.

Have a crypto agent and a news agent? Ask about "Bitcoin news" and both will contribute their expertise. The orchestrator handles the coordination so you get the best of all your agents in one response.

We've integrated LangGraph to give agents sophisticated multi-step reasoning capabilities. Agents now use the ReAct pattern (Reason → Act → Observe) to think through complex problems step by step.

We've launched a flexible credit-based billing system. Purchase credits when you need them, with no monthly commitments. Credits never expire, and you can enable auto-reload to keep your scheduled agents running smoothly.

Secure checkout powered by Stripe. View your usage history in the Usage page.

We've implemented intelligent prompt caching that dramatically reduces token usage for repeated or similar requests. The system automatically caches common prompt components and only sends unique parts to the LLM.

Upload your documents and let agents search through them using semantic search. Files are automatically chunked, embedded, and indexed so agents can find relevant information based on meaning, not just keywords.

We're excited to launch Telumin — your AI agent platform for autonomous intelligence. Deploy agents that monitor, analyze, and communicate insights 24/7, even while you sleep.