---
title: "8 Semantic Layer Tools for BI and AI Agents in 2027 | Dawiso Blog"
description: "Semantic layer tools for 2027 compared: dbt, Cube, AtScale, Snowflake, Databricks, Looker, Power BI, and the context layer that governs them for AI agents."
url: "https://www.dawiso.com/blog-post/best-semantic-layer-tools-2026"
language: "en"
---

# 8 Semantic Layer Tools for BI and AI Agents in 2027

Samuel Nagy

VP of Strategic Growth

Updated November 4, 2026  · Published June 24, 2026

A semantic layer is what lets a number mean the same thing in every dashboard and every AI agent. Going into 2027 the market still splits into four categories, but 2026 moved the pieces: the open standard for metric definitions found a neutral home, the warehouses started selling their own context for agents, and consolidation reached the category. The tools below cover the ones worth evaluating, and the twist for AI is still the same. Defining a metric is not the same as making it safe for an agent to act on.

TL;DR

Semantic layer tools fall into four categories: **standalone** (dbt, Cube, AtScale), **warehouse-native** (Snowflake, Databricks), **BI-native** (Looker, Power BI) and the **context layer** that governs the estate beneath them, a category Dawiso leads. In 2026 the open standard for metrics became Apache Ossie, Snowflake and Databricks began selling context for agents, and Fivetran and dbt Labs merged. Pick the engine that fits your stack, then govern the whole estate in a layer you own.

## What Changed Going Into 2027

When we first published this guide in June 2026, the four categories below already described the market. They still do. What moved during the year is who owns which piece, and how far each vendor reaches beyond its own product. Four changes matter if you are choosing a semantic layer for 2027.

**1. The open standard found a neutral home.** The Open Semantic Interchange (OSI), the vendor-neutral YAML format for metrics, dimensions and relationships launched in September 2025, [entered the Apache Incubator](https://ossie.apache.org/updates/ossie-enters-apache-incubator/) in 2026 and was renamed Apache Ossie. The specification did not change, and the coalition behind it has grown from 17 launch partners to more than 50 organizations. dbt Labs open-sourced MetricFlow as a reference implementation, and converters exist for dbt, Snowflake and Apache Polaris. Portable definitions are no longer a promise from one vendor, and that makes it cheaper to change engines later.

**2. The warehouses moved up the stack.** At its June 2026 Summit, Snowflake introduced [Horizon Context](https://www.snowflake.com/en/blog/horizon-context-governed-context/), a governed context layer inside Horizon Catalog that collects metadata, enriches it with business definitions and serves it to agents, with semantic views exposed over MCP. Two weeks later Databricks announced [Unity Catalog Metrics](https://www.databricks.com/blog/whats-new-unity-catalog-data-ai-summit-2026), Domains and Genie Ontology, and Microsoft extended Fabric IQ, which puts an ontology on top of Power BI semantic models. All three now describe their semantic layer as context for AI agents, and all three are strongest on data inside their own platform.

**3. Consolidation reached the category.** Fivetran and dbt Labs completed their all-stock merger on 1 June 2026 and now operate as one company that covers ingestion, transformation and metrics. For dbt users that means one vendor's roadmap across more of the stack, which is convenient until you want to move one piece.

**4. Analysts gave the wider layer a name.** On 20 October 2026 Gartner listed [AI context platforms among its top strategic technology trends for 2027](https://www.gartner.com/en/newsroom/press-releases/2026-10-20-gartner-identifies-the-top-strategic-technology-trends-for-2027) and predicts that by 2027, organizations that prioritize semantics in AI-ready data will increase agentic AI accuracy by up to 80% and reduce costs by up to 60%. Semantics are one part of that context, and it is the part this guide covers.

## What a Semantic Layer Is, and the Four Categories

A [semantic layer](https://www.dawiso.com/glossary/semantic-layer) sits between your raw tables and the people, or agents, asking questions. It holds the definitions. What "revenue" means, how "active customer" is calculated, which tables join to produce a metric. Without one, every dashboard and every AI agent re-derives those definitions on its own, and the answers drift apart. With one, a number means the same thing wherever it appears.

The landscape splits into four categories. Three of them focus on defining metrics: **standalone semantic layers** you run yourself, **warehouse-native layers** built into Snowflake and Databricks, and **BI-native layers** inside tools like Looker and Power BI. The fourth is less a new technology than a new name for an old discipline. The capabilities a **context layer** is built from, a data catalog, a business glossary, data lineage, and classification, have governed data for years. What changed is the consumer. Instead of serving that governed metadata to people through dashboards, a context layer assembles it and serves it to AI agents over an open protocol. The discipline is mature. Pointing it at AI is the new part.

The list below is grouped by those categories and numbered for reference, not ranked. The right pick depends on where your stack already lives, and most teams end up combining a semantic layer with a context layer on top.

One thing to hold onto while you read. Most of these tools now market themselves for AI, and several enforce access rules, define business meaning, and expose an MCP server of their own, so this is not a story about who has "context" and who does not. The real difference is scope. A semantic layer governs the metrics you model inside it. It does not catalog your whole data estate, classify sensitive data wherever it lives, trace lineage end to end across platforms, or bring several semantic layers together under one governed view. That wider, cross-platform job is what a context layer does, and it is the part worth owning whichever engine you pick.

## The 8 Tools at a Glance

The table summarizes where each tool lives and how far its definitions reach. "Reach" is the honest question for AI: a definition an agent cannot read outside one product does not stop that agent from guessing everywhere else.

| Tool | Category | Where definitions live | Reach beyond its own product |
| --- | --- | --- | --- |
| dbt Semantic Layer | Standalone | YAML next to dbt models (MetricFlow) | Portable through Apache Ossie, BI and agent integrations |
| Cube | Standalone | Cube data model, served over APIs | Warehouse-agnostic, sits in front of several sources |
| AtScale | Standalone | Universal semantic model with MDX | Serves Excel, Tableau, Power BI and AI tools |
| Dawiso | Context layer | Business glossary, generates semantic views through Ossie | Governs the estate across the platforms it connects to, serves agents over MCP |
| Snowflake | Warehouse-native | Semantic views in Horizon Catalog | Horizon Context reads external metadata (preview); definitions run in Snowflake |
| Databricks | Warehouse-native | Unity Catalog Metrics and metric views | Open source in Unity Catalog OSS; governance runs in Databricks |
| Looker | BI-native | LookML | Strongest inside Looker and Google Cloud |
| Power BI / Fabric | BI-native | DAX semantic models, Fabric IQ ontology | Strongest inside Microsoft 365 and Fabric |

## \1. dbt Semantic Layer

*Standalone. Best for teams who already model transformations in dbt.* The dbt Semantic Layer extends the dbt workflow into metric definitions, with MetricFlow underneath and definitions living in version-controlled YAML next to the models they describe. dbt Labs was a founding backer of the [Open Semantic Interchange](https://ossie.apache.org/updates/ossie-enters-apache-incubator/), now Apache Ossie, and open-sourced MetricFlow as a reference implementation, so definitions written in dbt stay portable across BI tools and AI agents rather than locked to one syntax.

The company behind it changed in 2026. Fivetran and dbt Labs completed their merger on 1 June 2026, so ingestion, transformation and metrics now sit with one vendor. For many teams that is a convenience. If you prefer each layer to stay swappable, it is a reason to keep the governance of those metrics in a layer that does not belong to the same company.

What dbt defines, it defines well, and that is the whole of its job: metrics, in code. It does not catalog the rest of your estate, classify sensitive fields, trace lineage across platforms, or surround a metric with the glossary, ownership, and provenance an agent needs. So dbt and Dawiso sit together rather than compete. dbt is the metrics engine your analytics engineers own; Dawiso catalogs those metrics alongside everything else, governs them, and serves them to any MCP-compatible agent, the setup in [context layer for dbt](https://www.dawiso.com/glossary/context-layer-for-dbt).

**Key capabilities**

- Metrics defined as code in version-controlled YAML (MetricFlow, now open source)
- Lives next to dbt transformations, in the same Git review
- Ossie-aligned, so definitions stay portable across tools

## \2. Cube

*Standalone. Best for headless analytics and AI agents.* Cube is a headless semantic layer. It exposes metrics over APIs (SQL, REST, GraphQL, and an AI-oriented interface) so embedded analytics, applications, and agents all query the same definitions. It is warehouse-agnostic and built to sit in front of more than one data source, which makes it a common pick when an agent needs consistent metrics without being tied to a single BI tool. The trade-off is operational: it is more engineering-heavy to stand up than a BI-native model, ships fewer pre-built BI-tool integrations than Looker or the dbt Semantic Layer, and adds self-hosting overhead unless you run Cube Cloud. It enforces its own access rules and ships an MCP server, so a catalog of your whole estate, cross-platform lineage, classification, and a business glossary are the wider job a context layer does above it.

**Key capabilities**

- Headless: metrics over SQL, REST, GraphQL, and AI APIs
- Warehouse-agnostic, sits in front of multiple sources
- Built for embedded analytics and AI agents

## \3. AtScale

*Standalone. Best for large enterprises with a mixed BI estate.* AtScale is a universal semantic layer with MDX support and aggregate-aware query acceleration, which is why enterprises with legacy OLAP commitments reach for it. Excel, Tableau, Power BI, and AI tools all hit one governed set of definitions over the same warehouse, and the acceleration layer keeps interactive queries fast as data volumes grow. It is priced for the enterprise, so smaller or early-stage teams tend to find it heavy, it carries more configuration overhead than headless options, and its most natural fit is organizations that already have an OLAP-cube heritage. AtScale governs the metrics it models, even across sources, and markets itself for AI; a catalog of your whole estate, a business glossary, classification, and end-to-end lineage are the wider job that sits in the context layer on top.

**Key capabilities**

- Universal semantic layer with MDX support
- Aggregate-aware query acceleration at scale
- Serves Excel, Tableau, Power BI, and AI tools alike

## \4. Dawiso

*The market leader for the context layer. It does the semantic-layer job, then goes past it.* Here the list bends. Dawiso does the work of a semantic layer and adds the layer on top. It does the semantic work itself: its [Business Glossary](https://www.dawiso.com/product/business-glossary) defines what your metrics and terms mean, each with one approved definition, an owner and a version history, and it generates governed semantic views through OSI, the vendor-neutral standard now called Apache Ossie, and writes them into Snowflake. Then it works at a scope a single metrics engine does not. A semantic layer governs the metrics you model in it, and the better ones enforce who may query them. Dawiso governs the whole estate around those metrics: it catalogs your assets across the platforms it connects to, classifies what is sensitive wherever it lives, traces lineage end to end, records who owns each definition, and sits above the semantic layers it connects to, such as dbt, Snowflake, Databricks, and Power BI, bringing their definitions into one governed view an agent can trust. That is the line we draw in [context layer vs semantic layer](https://www.dawiso.com/glossary/context-layer-vs-semantic-layer).

Dawiso reads metadata from more than 40 sources and builds one governed foundation across all of them: a [Data Catalog](https://www.dawiso.com/product/data-catalog) of what exists, the Business Glossary of what each term means, [classification](https://www.dawiso.com/glossary/data-classification) of what is sensitive, and [Interactive Data Lineage](https://www.dawiso.com/product/interactive-data-lineage) of where everything came from. Through the [Context Layer](https://www.dawiso.com/context-layer) and its [MCP Server](https://www.dawiso.com/product/mcp), it serves that governed context, definitions included, to any MCP-compatible assistant or agent, and each agent signs in as the person using it and gets only the tools that person's role allows. You can let Dawiso govern and serve a semantic layer it connects to, lean on its own glossary and semantic views, or do both.

One more reason it matters in 2027: Dawiso is independent of every warehouse and BI suite on this list. As Snowflake, Databricks and Microsoft each build context that works best on their own platform, the definitions, owners and lineage you build in Dawiso stay in a layer that does not change hands when you change engines.

**Key capabilities**

- Business glossary, catalog, lineage, and classification in one layer
- Generates governed semantic views through Apache Ossie (portable)
- Serves governed context to any MCP-compatible agent, scoped to the user's role
- Cross-platform, so context is not bound to one warehouse

"A semantic layer answers what the number is. A context layer also answers whether you can trust it."

[See it in action](https://www.dawiso.com/context-layer)

Dawiso Context Layer

Add governed context to your data stack and serve it to any MCP-compatible agent.

## \5. Snowflake (Semantic Views and Horizon Context)

*Warehouse-native. Best for teams standardized on Snowflake.* [Semantic views](https://www.dawiso.com/glossary/snowflake-semantic-views) are Snowflake's governed business-logic layer, and [Cortex](https://www.dawiso.com/glossary/snowflake-cortex) agents answer natural-language questions by reading them rather than raw tables. In 2026 Snowflake added a lot around them. Semantic View Autopilot drafts semantic views from existing SQL, Tableau and Power BI files, semantic views can be exposed to agents over MCP, and Horizon Context, introduced in June 2026, turns Horizon Catalog into a context layer that collects metadata, adds business definitions and serves them to AI and BI. In October Snowflake put metadata connectors for 18 external sources, including Power BI, Tableau, dbt and Databricks, into public preview, so Horizon now reads metadata from outside Snowflake as well. Most of these pieces are still in preview, so check status before you plan on them.

It also pairs unusually well with Dawiso, and in both directions. Dawiso scans Cortex agents and their semantic views into interactive lineage, so you can see exactly which agent reads which table, and it generates governed semantic views back into Snowflake through Ossie. We walk through that two-way setup in [Govern Snowflake Cortex Agents and Semantic Views](https://www.dawiso.com/blog-post/snowflake-cortex-agents-semantic-views-dawiso). The trade-off is where the governance lives. Horizon Context can read metadata from other platforms, but the definitions, policies and agents it governs run in Snowflake, so once Snowflake is one of several platforms, a context layer you own holds the single cross-platform definition, the role we describe in [context layer for Snowflake](https://www.dawiso.com/glossary/context-layer-for-snowflake).

**Key capabilities**

- Semantic views read by Cortex agents, exposable over MCP
- Horizon Context with external metadata connectors (preview)
- Two-way with Dawiso: scanned to lineage, Ossie views written back

## \6. Databricks (Unity Catalog Metrics and Genie)

*Warehouse-native. Best for lakehouse teams on Databricks.* Metric views in [Unity Catalog](https://www.dawiso.com/glossary/databricks-unity-catalog) define metrics once, and [AI/BI Genie](https://www.dawiso.com/glossary/databricks-ai-bi-genie) reads that semantic metadata to answer questions, with Unity Catalog's row-level security and column masking flowing into the answer. At Data + AI Summit 2026 Databricks went further with Unity Catalog Metrics, which adds multi-fact relationships, level-of-detail calculations and import from Power BI and Tableau, with several pieces in preview or beta, plus Domains and a Genie Ontology it describes as a continuously learned context layer. Metrics is open source in Unity Catalog OSS and aligned with Apache Ossie. The governance is real and, as with Snowflake, the scope is the platform.

Dawiso works alongside it rather than against it. It scans Unity Catalog and Genie into interactive lineage and layers a cross-platform glossary, classification, and governance on top, then serves the result to any agent over MCP, the setup we describe in [context layer for Databricks](https://www.dawiso.com/glossary/context-layer-for-databricks). Unity Catalog is itself a catalog and governance layer, so it overlaps a dedicated governance tool more than the pure semantic layers do. The difference is reach. It governs within Databricks, while a context layer holds one definition across Databricks, your warehouse, and your BI tools, and reaches agents running outside the lakehouse.

**Key capabilities**

- Metric views and Unity Catalog Metrics defined once
- Genie answers with row-level security and masking applied
- Works alongside Dawiso for cross-platform reach

## \7. Looker (LookML)

*BI-native. Best for organizations on Google's stack.* LookML has been a semantic model for more than a decade, and Google has extended it with Gemini so users can ask questions in natural language against governed metrics. It is mature and trusted for BI, with the trade-off common to BI-native layers: the model is strongest inside Looker and the wider Google ecosystem, and less portable to tools and agents outside it, a BI-tool lock-in worth weighing. For Microsoft-aligned teams, Power BI semantic models fill the same BI-native bracket.

**Key capabilities**

- Mature LookML semantic model, more than a decade old
- Natural-language questions through Gemini
- Strongest inside Looker and Google's ecosystem

## \8. Microsoft Power BI / Fabric

*BI-native. Best for Microsoft-centric organizations.* Power BI semantic models, the DAX-based layer once called datasets, are among the most widely deployed metric definitions anywhere, now part of Microsoft Fabric and surfaced to Copilot. In 2026 Microsoft built Fabric IQ on top of them: the semantic model supplies the measures, and an ontology adds business entities and relationships that Fabric data agents, Copilot Studio and MCP clients can read. Fabric IQ is generally available, while the ontology itself was still labeled preview in Microsoft's documentation at the time of writing. For teams already in Microsoft 365 and Fabric, it is the path of least resistance. As with Looker, the definitions are most at home inside the Microsoft estate, and a layer above helps when agents need to read them alongside data from everywhere else.

**Key capabilities**

- DAX-based semantic models, among the most widely deployed anywhere
- Fabric IQ ontology on top of the semantic model
- Best fit for Microsoft 365 and Fabric estates

## How to Choose

Two questions settle most of this. First, where does your stack already live? If you are deep in one platform, its native layer is the fastest start, and in 2026 those native layers got noticeably better. If you run several, a standalone or BI-agnostic layer keeps you from redefining the same metric in each one, and Apache Ossie makes it easier to move definitions between them.

Second, and the one that matters most for AI: how will agents read these definitions safely? A semantic layer hands an agent a governed number, and many enforce who may query it. What it does not hand over is the wider context around that number: a catalog of the rest of your estate, end-to-end lineage, classification of what is sensitive across platforms, and a record of who owns each definition. An agent that acts beyond those metrics without that context is a liability. Pair your semantic layer with a context layer, and expose it to agents over [MCP](https://www.dawiso.com/glossary/model-context-protocol-mcp).

There is an architectural reason to keep that context layer separate, and a commercial one. 2026 showed how fast ownership moves in this market: an ingestion vendor merged with the leading transformation tool, and every warehouse now sells its own context. A composable stack, where each layer is a best-of-breed tool you can swap, keeps your switching costs low. If a warehouse's pricing climbs, you can move that one layer without re-governing your whole business, because your definitions, ownership, and lineage live in a layer you own rather than inside the vendor you are leaving.

## Where Dawiso Fits

Dawiso is that context layer, and on product it is one of the two market leaders in the category, alongside Atlan, at a lower price. It reads metadata from the warehouses, BI tools, and transformation layers where your data and metrics already live, including Snowflake, Databricks, Power BI, and dbt from this list. Across all of them it builds one governed foundation, a catalog of what exists, a business glossary of what each term means, classification of what is sensitive, and end-to-end lineage, and serves it to any MCP-compatible agent. Dawiso is where the meaning, ownership, and provenance of those metrics live, and it keeps that context yours to move. For the wider market of tools that do this job, see [8 context layer tools for AI agents in 2027](https://www.dawiso.com/blog-post/best-context-layer-tools).

The semantic layer answers what the number is. Dawiso answers whether you can trust it, and lets every agent read the same trusted answer across the sources it governs.

## FAQ

**What is a semantic layer, and why do AI agents need one?**

A semantic layer is a curated description of your data in business terms: what each metric means, how it is calculated, and which tables and joins are correct. AI agents need one because without it, every agent re-derives those definitions from raw column names and they drift apart, so the same question returns different answers in different tools. With a shared semantic layer, a number means the same thing in every dashboard and every agent.

**What are the best semantic layer tools for BI and AI agents in 2027?**

The market splits into four categories. Standalone layers you run yourself (dbt Semantic Layer, Cube, AtScale), warehouse-native layers (Snowflake semantic views with Horizon Context, Databricks Unity Catalog Metrics with Genie), BI-native layers (Looker and LookML, Microsoft Power BI with Fabric IQ), and the context layer that governs the estate beneath them (Dawiso). The right pick depends on where your stack already lives, and most teams pair a semantic layer with a context layer for AI.

**What changed in the semantic layer market in 2026?**

Three things. The Open Semantic Interchange (OSI) entered the Apache Incubator and was renamed Apache Ossie, so portable metric definitions now have a neutral home. The warehouses moved up the stack: Snowflake introduced Horizon Context and Databricks introduced Unity Catalog Metrics and Genie Ontology, both pitched as context for agents. And consolidation reached the category when Fivetran and dbt Labs completed their merger on 1 June 2026.

**What is the difference between a semantic layer and a context layer?**

A semantic layer defines what a metric means and often enforces who may query it. A context layer works at a wider scope: it catalogs your whole data estate, classifies sensitive data wherever it lives, traces lineage end to end, records ownership, and can sit above the semantic layers it connects to, then serves all of it to agents over an open protocol. The semantic layer answers what the number is. The context layer answers whether you can trust it across everything you run.

**Do I need both a semantic layer and a context layer?**

For BI inside one tool, a semantic layer is often enough. For AI agents that act across your data, you need both. A semantic layer governs the metrics you model in it, but it does not catalog the rest of your estate, classify sensitive data across it, or trace lineage end to end, and an agent acting beyond those metrics without that context is a liability. Keeping the context layer separate also keeps your stack composable, so you can swap a warehouse without re-governing the business.

**Is Dawiso a semantic layer?**

Dawiso does the semantic-layer job (its Business Glossary defines metrics and terms, and it generates governed semantic views through the open OSI standard, now Apache Ossie) and goes beyond it. Its role is the context layer on top: it catalogs, classifies, and traces lineage across the platforms it connects to and serves governed context to any MCP-compatible agent. You can run Dawiso alongside a dedicated semantic layer like dbt or Snowflake, or use its own glossary and semantic views.

## Sources

- Apache Ossie - [Apache Ossie (Incubating): The New Name for Open Semantic Interchange](https://ossie.apache.org/updates/ossie-enters-apache-incubator/) (July 10, 2026)
- Snowflake - [Snowflake Horizon Context: The Governed Context Layer for AI, BI and Apps](https://www.snowflake.com/en/blog/horizon-context-governed-context/) (June 2, 2026)
- Databricks - [What's new with Unity Catalog at Data + AI Summit 2026](https://www.databricks.com/blog/whats-new-unity-catalog-data-ai-summit-2026) (June 16, 2026)
- TechTarget - [Fivetran, dbt Labs complete merger to form data layer for AI](https://www.techtarget.com/data-technologies/news/366643590/Fivetran-DBT-Labs-complete-merger-to-form-data-layer-for-AI) (June 1, 2026)
- Gartner - [Gartner Identifies the Top Strategic Technology Trends for 2027](https://www.gartner.com/en/newsroom/press-releases/2026-10-20-gartner-identifies-the-top-strategic-technology-trends-for-2027) (October 20, 2026)
- Microsoft Learn - Fabric IQ and ontology (preview) documentation

### Table of contents

What Changed Going Into 2027What a Semantic Layer Is, and the Four CategoriesThe 8 Tools at a Glance1. dbt Semantic Layer2. Cube3. AtScale4. Dawiso5. Snowflake6. Databricks7. Looker8. Microsoft Power BI / FabricHow to ChooseWhere Dawiso FitsFAQSources

## More like this

Keep reading and take a deeper dive into our most recent content on metadata management and beyond:

### [Why Text-to-SQL Breaks in the Enterprise](https://www.dawiso.com/blog-post/why-text-to-sql-breaks-in-the-enterprise)

Read more

### [Data Catalog Tools for 2027](https://www.dawiso.com/blog-post/best-data-catalog-tools-2026)

Read more

### [8 Context Layer Tools for AI Agents in 2027](https://www.dawiso.com/blog-post/best-context-layer-tools)

Read more