> ## Site Index
> Fetch the site index at: https://www.credibledata.com/llms.txt
> Use this file to discover all available pages before exploring further.

# Credible vs. TextQL

> TextQL is an AI analyst with its ontology in your own repository. Credible serves an open model to whatever agent your team already runs.

## How TextQL describes itself

TextQL is an enterprise AI data analyst. Ana writes SQL, runs Python, searches the web and produces charts, reports and files, working from an ontology — metric definitions, entity relationships, joins and access rules authored as .tql files in a git-backed repository — across more than 50 connected sources. Roughly half of its production workloads run on-premises or in customer VPCs, alongside a purpose-built warehouse that runs in the same place.

## Credible and TextQL compared

| Dimension | Credible | TextQL |
| --- | --- | --- |
| Category | The AI Analytics Engine — you write down what data means, and it generates the stack under it | AI data analyst over the whole stack, with an ontology layer and a warehouse of its own |
| Who it is built for | The whole organization, not one central team — ops, finance and product; spreadsheet users through ML engineers | Enterprise data teams, and the business users whose questions queue behind them |
| Primary interface | The agent you already use, over MCP — plus Credible Workspaces, dashboards, notebooks and reports for the people who want a UI. Every one of them reads the same model, and the model lives in your repository | Chat — threads with Ana in the TextQL app, in Slack, and from any MCP client |
| Modeling language | Malloy — a modern programming language for data: imports, inheritance, public and private members, and queries that compose into new sources | .tql — a SQL-native semantic layer format, kept as reviewable files in a git-backed ontology repository |
| What is open | Open core. Malloy is open source, and we build and maintain Malloy Publisher — the open-source server for Malloy models, and where our agent skills and MCP retrieval tools are open source too, runnable in any agent with no Credible account. Your model is code in your repository, and Credible is in the Apache Ossie ecosystem for semantic interchange | Vendor-owned format, with the ontology itself kept as files in a repository you control |
| How the model gets built | Model first: you write down what data means, and the engine derives the pipelines, storage and serving from it. Open-source MCP tools and agent skills help you capture that meaning from where it already lives, running in your coding agent | Ana maps relationships across the connected sources, and .tql files are authored and reviewed on top |
| Governance | One annotation on the source, in the same file as the logic it governs — no grant objects to declare, no rule to re-attach on every surface that exposes the data, no attribute table kept in step by hand. A row filter or an authorize gate over attributes the server resolves from verified identity, and everything reading that source inherits it: dashboards, notebooks and agents alike | Role-based access control with SSO and SCIM, and role and tenant rules declared in the ontology |
| Materialization and caching | One annotation on the source, in the same file as the logic it materializes — no derived-table block, no rollup definitions, no refresh triggers, no orchestration run to schedule. Optional per source: query your own warehouse directly, or hold a source hot in Credible’s own storage, which is how you get fast serving without buying or banging your head against a warehouse to get it | A purpose-built warehouse running inside your environment, so exploratory work over raw data does not wait on modeling |
| How an agent finds the right data | Search. Typed targets — source, dimension, measure, view, even a dimensional value — are matched against an index of the model and come back ranked, so an agent asks for what it needs instead of picking a model and touring it | Ana works from the ontology and its connectors, and picks its own tools per question rather than walking a fixed sequence |
| Where the model can be used | One model served to every surface — agents over MCP, plus APIs, an SDK, embedded dashboards, notebooks and HTML data apps. Results carry an interactive UI resource (MCP Apps, the official extension), so a client that supports it renders a real chart or table instead of the model re-narrating rows | Threads in the app, Slack, playbooks scheduled to email, embeds, a v2 REST API, and Ana itself as an MCP server |
| Scale and portability | Globally distributed, high-availability serving that scales horizontally — workers scale out under load rather than being sized in advance. Warehouse-agnostic on open-source Malloy, and the model is portable because the language is | Over 50 connectors — warehouses, BI tools and SaaS APIs — deployed as SaaS, in your VPC, or on-premises by Helm chart |
| What you pay for | Usage, not seats — unlimited users on every plan, so adding people never changes the bill. Three meters, pooled per organization and starting free: tokens, compute time and storage. Your agent’s own LLM tokens are never billed to you, and a query that runs on your own warehouse is not metered for the compute — only for the result it hands back | A tiered subscription carrying a monthly allowance of Agent Compute Units, with overage beyond it and no per-seat charge |

## Where Credible takes a different approach

### What the extra compute is for

Both of us bring compute beyond the model. TextQL runs a purpose-built warehouse inside your environment, and that is what lets Ana work over raw, unprepared data instead of waiting on the months of modeling that usually come first — a direct answer to a real problem. Credible’s serving layer is pointed at a different job: it materializes modeled sources, indexes them and holds them hot in memory, so retrieval and queries come back in milliseconds rather than warehouse-scan minutes. It is also optional, and decided per source: anything you would rather leave where it is gets queried there instead, across BigQuery, Snowflake, Postgres, MySQL, Databricks, Trino and DuckDB. That matters most to a team running analytics on Postgres who wants a few views served fast without buying a warehouse to get it.

### SQL fragments, and a modeling language

.tql is deliberately legible: authored SQL templates with sanitized interpolation and typed parameters, kept as files rather than compiled behind a planner, so a reviewer — or a model — can read exactly what will run. Malloy takes the other route and gets composition for it: imports, inheritance, and public and private members, so a definition is written once and reused, and the compiler emits SQL for seven dialects. Both end up as code in a repository. The difference shows up as a model grows: a .tql file states each surface explicitly, where Malloy lets one definition stand behind many, which is what keeps a large model something a person can still hold in their head and an agent can extend without breaking.

### Open files, and an open language

Keeping the ontology as files in a repository you control is a good decision, and not every vendor in this category makes it. Credible carries the same idea one step further out: Malloy is an open-source language, Malloy Publisher is the open-source server we build and maintain for it, and Credible is in the Apache Ossie ecosystem — so the model is readable and runnable by tools that are not ours.

### An analyst you call, and context your agent reads

Ana is the analyst. It plans the work, writes the SQL, runs the Python and hands back the result, and it is reachable from Cursor, Claude, ChatGPT and other MCP clients as well as from TextQL and Slack. Credible serves the model rather than an agent: sources, dimensions, measures, the rules attached to them and the engine underneath, over MCP — so the reasoning happens inside the agent your team already runs, with results carrying an interactive UI resource so a chart renders in that client. Both put an answer in front of someone; they differ on whose agent does the thinking.

Capability claims about TextQL were checked against public sources in August 2026. Sources: Ana, the ontology and the .tql format — TextQL documentation; Connectors, deployment types, RBAC and Ana as an MCP server — TextQL documentation; $17M round anchored by Blackstone Innovations Investments — TextQL, 2026.