Maxim AI release notes www.getmaxim.ai

Fix

• Adds handling for network resets during long polls.

Improvement

• Accommodates new changes for LangGraph handlers.

// pip
pip install maxim-py==3.4.11
// UV
uv add maxim-py==3.4.11

New

• New cookbook to use dSPY evaluators using Maxim SDK

New

Whether you're building dynamic prompts, injecting structured context, or running personalized prompt tests at scale—Jinja2 syntax makes it easier than ever to craft powerful, composable prompts right inside our Prompt playground.

With Jinja2 support, you can structure your prompts like actual code—loop through lists, add conditional statements, and dynamically inject variables—all while keeping your prompt logic clean and reusable. Follow this demo to learn more.

New

Obsessed with Grok? As a step towards becoming your one-stop solution for all experimentation and evaluation needs, we’ve added support for xAI models—giving you a greater selection pool and flexibility to test, compare, and refine your AI workflows.

No complex setup—just plug in your API key and start building in minutes.

New

Google's latest Gemini 2.5 Pro model is now available on Maxim. Leverage its advanced reasoning and multimodal capabilities to design custom evaluators and prompt experiments.

Start using this model via the Google provider:
✅ Go to Settings > Models > Google and add Gemini 2.5 Pro Experimental

New

Tired of managing your secrets across multiple places? Vault is a centralized, encrypted home for all your sensitive variables—from API keys to tokens. Use your secrets in Maxim workflows or API evals by simply referencing them from Vault.

All secrets are encrypted at rest and in transit using RSA 2048-bit encryption, ensuring that even if someone gains access to the system, your data remains protected without the private key. Vault is built for robust security and peace of mind.

New

While our SDK support has made it seamless to work with Maxim, we’ve now introduced a comprehensive REST API—giving you even more flexibility to integrate Maxim into your existing workflows. Use Maxim APIs to:

• Experiment with prompts, organize them, and programmatically create and deploy prompt versions.
• Configure alerts and integrate with tools like Slack and PagerDuty to stay on top of prod issues
• Manage datasets at a granular level—perform CRUD operations on entries, columns, and datasets, or create data splits directly via endpoint.
• Create and manage log repos, and fetch logs without needing to access the UI.

New

Teams today spend countless hours manually going back and forth with their agents to assess quality and identify failure modes. And it’s not a one-time thing— unintended regressions happen all the time, whether due to model updates or optimizations based on user feedback.

Introducing Maxim’s agent simulation! Now, you can evaluate your AI agents in just three simple steps:

1️⃣ Define the real-world scenarios, user personas, and any other context you want to test your agents on.
2️⃣ Pick the right evaluators, from predefined evaluators to custom metrics or human reviews, for your use case and trigger a test run.
3️⃣ Analyze your agent’s performance, debug issues, and iterate.

The best part? Simply bring your agents via an API endpoint and get started in minutes!

New

Teams store sensitive customer data, API keys, and production logs on the Maxim platform—and we take that trust seriously. To enhance account security, Maxim now supports two-factor authentication (2FA), adding an extra layer of protection against unauthorized access to your data.

Enable 2FA for all users by simply navigating to:

Settings ➡️ Organization info ➡️ Two-factor authentication.

New

Our new Prompt Diff feature lets you compare two prompt versions side-by-side with a Git diff-style view revealing every tweak. See how your prompt is evolving over time and correlate changes in performance with prompt changes.

New

Need help? With our in-app support, you can get assistance, ask questions, and share feedback — all with a click of a button. With 24/7 support, we’re here to ensure you’re never blocked and can stay focused on building, testing, and scaling your AI workflows.