Home

Bloom Labs » Home
★ INTERNAL MILESTONE: Homan 2.96 correctly multiplied 3-digit numbers 2 out of 3 times in-house — a Bloom Labs first! ★ Bloom 4.6 completes internal benchmarking -- pricing & access details still to come ★ BabyMax 5.76 now under internal safety review ahead of any public release ★ Read our new Safety & Responsible Scaling whitepaper ★ Bloom Labs featured in Web 2.0 Weekly Digest

Intelligence, Blooming.

Bloom Labs builds friendly, capable artificial intelligence for the next generation of the connected web. Meet Homan, Bloom, and BabyMax — our growing family of models, built for researchers, businesses, and everyday people alike.

Explore Our Models » Read Our Research

Welcome to Bloom Labs

Bloom Labs, Inc. is pushing the bleeding edge of artificial intelligence research. Our labs are home to some of the most advanced neural architectures ever trained — massively parallel, next-generation systems engineered to reason, write, and problem-solve at a level the web has never seen before.

Our flagship research produces a family of cutting-edge large language models — Homan, Bloom, and BabyMax — each representing the absolute state-of-the-art for its size class. Built on proprietary training breakthroughs and tuned for a different balance of raw intelligence, speed, and cost, this is the most advanced model family Bloom Labs has ever shipped.

And we're just getting started. Every one of these next-generation systems is developed under our safety-first research program, so the most powerful AI we've ever built is also the most responsible.

Our Model Family

Updated for 2026
Flagship

Homan 2.96

Our most capable model. State-of-the-art reasoning, coding, and multimodal understanding for the toughest problems on the web.

Status: Coming Soon — internal testing only

Learn More »
Balanced

Bloom 4.6

The sweet spot of speed, cost, and intelligence. Perfect for chatbots, customer support, and everyday web applications.

Status: Coming Soon — internal testing only

Learn More »
Lightweight

BabyMax 5.76

Small, speedy, and surprisingly smart. Built for instant replies, mobile widgets, and high-volume workloads.

Status: Coming Soon — internal testing only

Learn More »

Latest From The Research Blog

View All »

Homan 2.96 Benchmark Report: Reasoning, Code, and the Curious Case of Multiplication

We are beyond proud to report Homan 2.96's full benchmark results — including a landmark first for Bloom Labs: correctly multiplying two 3-digit numbers, two times out of three. Full breakdown inside.

Read more »

Introducing Homan 2.96: Our Most Capable Model Yet

Homan 2.96 has completed internal development as the newest and most capable member of our model family. Here's what's new, and what it will mean for developers once it's released...

Read more »

Constitutional Alignment: How We Keep Bloom Labs Models Safe

Safety isn't a feature we bolt on at the end — it's baked into how we train every model. A look inside our alignment research...

Read more »

Scaling Laws Revisited: What We Learned Training Bloom 4.6

Bigger isn't always better. Some notes from the trenches on compute-optimal training for our mid-size model...

Read more »

You Are Visitor Number

Thanks for stopping by Bloom Labs! Don't forget to sign our guestbook and add us to your favorites.