# Dex makes agents better analytics engineers.

> On dbt Labs' 75-task analytics-engineering benchmark, Dex + Claude Sonnet 5 resolves more tasks than any other run we measured, and beats dbt's own published agent skills.

- Benchmark: [ADE-bench](https://www.exmergo.com/benchmarks/ade-bench) (Analytics Engineering Benchmark), published by dbt Labs
- Run date: 2026-07-21
- Headline result: 57 / 75 tasks resolved (76.0%) with Claude Sonnet 5
- Benchmark repository: https://github.com/dbt-labs/ade-bench

## How we ran it

ADE-bench hands an agent a dbt project on DuckDB and asks it to fix a broken model, build a new one, or extend the semantic layer, then scores whether the project's tests pass.

- **Tasks**: 75 tasks across 8 domains (airbnb, f1, asana, intercom, quickbooks, helixops_saas, analytics_engineering)
- **Agent**: Claude Code, one attempt per task, up to 50 episodes
- **Dex**: Supplied as the exmergo/dex skill plugin
- **Baseline**: Identical setup with no plugin

## Results

| Run | Resolved | Accuracy | Cost |
| --- | --- | --- | --- |
| Claude Sonnet 5 (with dex) **[best]** | 57 / 75 | 76.0% | $35.95 |
| Claude Fable 5 (with dex) | 56 / 75 | 74.7% | $91.98 |
| Claude Opus 4.8 (with dex) | 54 / 75 | 72.0% | $43.38 |
| Claude Sonnet 5 (baseline) | 53 / 75 | 70.7% | $30.76 |

Dex + Sonnet 5 leads at 76%, for about 2.5x less than Fable 5 and ~17% less than Opus 4.8. With Dex, accuracy holds in a 72–76% band across all three models while cost ranges from $36 to $92, so the practical call is to run an inexpensive model.

For context, dbt's published agent skills reported 58% on this benchmark with Opus 4.6.

## Reading the numbers

- These are single-run results (one attempt per task), so treat small gaps between runs as noise.
- The raw results.json for every run is committed under experiments/, alongside the harness configuration.

Measured by Exmergo on dbt Labs’ ADE-bench. Canonical page: https://www.exmergo.com/benchmarks/ade-bench/2026-07-21