Hex

Hex is an AI analytics platform for data teams and business users working with company data. It brings notebooks, self-serve questions, and interactive data apps into one workspace. Its context engine grounds AI answers in warehouse data, semantic models, business rules, and analyst logic.

Key Highlights

• Ask questions in Hex, Slack, or via MCP with shared context
• Build SQL and Python analyses with the Notebook Agent
• Publish notebooks as apps, dashboards, reports, or presentations
• Govern agents with Context Studio, observability, endorsements, and warnings
• Connect warehouses, dbt metadata, OAuth database access, Git exports, orchestration, and the API

What Makes It Different

Hex's strongest idea is shared context. Database descriptions, semantic models, rules, and reusable logic feed the agents, so a Slack question and a notebook investigation use the same definitions.

A Thread can become notebook work or an app without rebuilding logic elsewhere.

Features & Capabilities

Analysts work with SQL, Python, charts, app builder controls, and AI help for query writing, debugging, visualization, and multi-step analysis. Threads gives non-technical users a plain-language path to answers without SQL or Python.

For the data stack, Hex connects to warehouses, lakehouses, databases, dbt docs and metrics, OAuth database connections, and GitHub or GitLab exports. Teams can run projects from Airflow, Dagster, or Prefect and extend workflows through the API.

User Ratings and Testimonials

Hex is rated on G2 as an industry leader, but no average rating is published. Customer quotes praise self-serve data access, semantic model authoring, app scaffolding, and SQL debugging. Practical limits are plan-based: Community caps notebooks, Professional caps published apps, and advanced compute is for Team and Enterprise.

Pricing & Value

• Community: Free, with a notebook agent trial, any data source, up to 5 notebooks, all cell types, and small compute
• Professional: $36 per Editor/month, with Notebook Agent, standard credits, unlimited notebooks, 5 published apps, 30-day history, and medium compute
• Team: $75 per Editor/month, with Threads Agent, Semantic Model Agent, extended credits, unlimited apps, scheduled runs, alerts, shared components, and advanced compute add-ons
• Enterprise: Custom pricing, with Explorer seat add-on, audit logs, premium credits, OAuth database connections, Observability API, OIDC SSO, and optional HIPAA, embedded analytics, single tenant, or custom Docker image add-ons

Extra small, small, and medium compute profiles are free. Team and Enterprise can use pay-as-you-go advanced compute, with large profiles from $0.32/hour and GPU profiles from $2.93/hour, billed per minute.