Brim January 2026 Release: Structured Data, Project-Level LLM Configuration, and a Smoother Experience
January 29, 2026

We’re kicking off 2026 with a release focused on something we’ve heard again and again:
Real-world workflows need both structured and unstructured data, and the flexibility to abstract with the right LLM model.
Your teams work across structured datasets, unstructured notes, multiple abstraction strategies, and different performance and cost needs, often all within the same organization. This January release makes Brim more flexible, more configurable, and easier to use, so it fits the way your teams actually work.
🧩 Structured Data: Bring It All Together
Many abstraction workflows rely on a mix of structured data (like demographics, labs, or coded values) and unstructured data (like clinical notes). Until now, those workflows often lived in separate tools and pipelines.
With this release, Brim now supports structured data end-to-end.
You can upload structured data such as demographics, lab values, or other tabular inputs and use it directly alongside abstracted variables to:
- Use structured data as input to abstracted variables
- Filter during generation and review using structured values
- Reduce unnecessary generation and focus effort where it matters.
- Export structured and abstracted data together
- Produce cleaner, more complete datasets downstream.
This makes it easier to combine known facts with AI-powered abstraction, improving accuracy, reducing cost, and producing more useful outputs.
To get started, go to Project Setup → Project Data → Add Data.
Read more in our documentation.
🤖 Configure LLMs by Project
Different projects have different goals. Some need maximum accuracy. Others prioritize speed, cost efficiency, or experimentation. Until now, Brim required you to choose one default LLM and one Advanced LLM for your entire workspace.
That changes today.
Brim now lets you configure multiple LLMs and assign them at the project level.
You can:
- Customize compute usage by project
- Optimize performance and cost depending on workload.
- Experiment with different models on the same data
- Compare outputs and refine workflows quickly.
- Balance cost, speed, and quality per use case
- Match model choice to task complexity.
To get started, go to Settings → Admin → LLM Config
Read more in our documentation.
🎨 Design & Usability Improvements
We’ve made a series of design and interaction improvements across Brim to make everyday workflows smoother and more pleasant.
These include:
- Improved contrast for better readability
- More consistent interaction patterns
- Helpful icons and visual cues
There are no new concepts to learn, just a more polished experience that stays out of your way and helps you move faster.
You can learn more in our Accessibility Statement.
✨ How to Get the Latest Brim Version
This release version is: 2026.01.28.
You can check your current version in Settings → About.