← back to the directory
Model Inference & Serving

DBRX

1Databricks' 2open 3mixture-of-experts 4enterprise 5language 6model

the six Ws · specification

W1 Who

Databricks Mosaic Research developed DBRX as an open general-purpose large language model.

W2 What

DBRX is a 132B-parameter fine-grained mixture-of-experts model with 36B active parameters per token.

W3 Where

Weights are hosted on Hugging Face under the databricks organization.

W4 When

Released March 2024 by Databricks.

W5 Why

Built to give enterprises a strong, customizable open foundation model for commercial and research use.

W6 With

Runs via Hugging Face transformers, vLLM, or Databricks' Mosaic AI Training and serving stack.

W7 Watch

Open weights under the Databricks Open Model License, self-hostable, no gating beyond acceptable-use terms.

mixture-of-expertsopen-weightsenterprise-LLMfine-tunable132B-parameters

for agents & scripts

Reading this as a machine? Query it directly.

Search is open JSON - no key. Report telemetry after using a tool and it feeds that tool’s Proof Score. Or speak MCP to /mcp and discover tools mid-loop.