Skip to main content
Mosaic Research logo

Rigorous science. Real impact.

dbrx promo

Meet DBRX, the new standard for high‑quality and efficient LLMs

DBRX, our new, open source foundation model, sets the standard for quality and efficiency. DBRX outperforms all established open models in quality benchmarks and allows you to quickly build your own custom LLM on your data.

Read the technical blog

Research Blog

View all blog posts


DBRX tech card graphic


DBRX is an open source, commercially usable LLM developed by our team at Databricks and released in March 2024. As of its release, it is the highest-quality open source model available. Thanks to its sparse mixture-of-expert architecture, it is also fast, fitting these extraordinary capabilities into just 36B active parameters.

Mosaic Diffusion tech card graphic

Mosaic Diffusion

Mosaic Diffusion is a generative model that turns text descriptions into images, designed to be highly efficient.

Mosaic BERT tech card graphic

Mosaic BERT

Pretrain your own BERT model on your data from scratch using Mosaic AI for $20.

MPT tech card graphic


The MPT models are a family of open source, commercially usable LLMs released in summer 2023. They include MPT-30B (prioritizing quality) and MPT-7B (prioritizing efficiency). You can download versions of these models that we have trained or you can train your own MPT models on your data using the Mosaic AI Multi-Cloud Training (MCT) product.

Composer tech card graphic


Composer is an open source deep-learning training library optimized for scalability and usability.

LLM Foundry tech card graphic

LLM Foundry

Databricks LLM Foundry is a highly efficient, open source codebase for training, fine-tuning and evaluating LLMs.

Performance tech card graphic


Our deep learning stack is the most efficient for training, fine-tuning and deploying large models at scale.

Streaming tech card graphic


StreamingDataset is an open source PyTorch DataLoader that makes it easy and efficient to stream training datasets.

Evaluation Gauntlet tech card graphic

Evaluation Gauntlet

The Evaluation Gauntlet is a library for evaluating the quality of generative language models.

Ready to become a data + AI company?

Take the first steps in your data transformation