Blog

Introducing DataBench

A frontier benchmark for complex data work and analytical reasoning

Introducing Evals

Measure Hex Agent performance and test context changes safely before they ship, all from the CLI.

Trust used to be implied

Evals aren't just how you fix your agent — they're how you convince people to trust it.

Caching is all you need

Our cache hit rate looked great. Our cost data disagreed. Here's what we found when we stopped trusting the average.

Accuracy and AI analytics: How the best data teams obsess over context

AI analytics depends on context for trusted answers. Here's how data teams at Ramp, Underdog, and Hover build and manage that context with Hex.

The life and times of data teams in the AI era

When everyone is a data person, what does it mean to be a data team?

Building semantic models in Hex: A playbook for governed & trustworthy metrics

A practical build recipe for Hex admins and data leads

Credits and usage visibility for Hex Agents

We’re rolling out monthly credit grants for Hex Agents, plus new ways to monitor usage and buy add-on credits.

Kimi K2.7 in Hex: near-frontier analytics at a fraction of the cost

Open models have finally caught up to their closed counterparts

Fed for a day

AI didn't change who's hungry, just who gets to fish.

Introducing Agent Tasks

Schedule the Hex Agent to answer your most important recurring questions automatically — and deliver results straight to Slack or email

Bring the web to your data

Hex has always been able to search your company’s data, now search the web.

Are Semantic Layers the Cornerstone for AI Analytics?

Are they the secret to trusted, accurate answers from data? The people selling them would certainly say so

We had to build new evals for Fable

Claude Fable 5 is the first model since Opus 4.5 to meaningfully improve at analytical reasoning

Hex is now in Codex

You can bring Hex directly into Codex to run analyses, pull in existing work, and spin up new projects.