Dex makes agents better analytics engineers.
Accuracy on ADE-bench
0.0%
0 / 75 tasks resolved
Dex + Claude Sonnet 5 on dbt Labs’ ADE-bench.
Results
Each run is one attempt per task across the full 75-task suite.
RunResolvedAccuracyCost
Claude Sonnet 5with dex
Best resultClaude Fable 5with dex
Claude Opus 4.8with dex
Claude Sonnet 5baseline
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.
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
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.
Ready to turn maintenance into an automated habit?
Install the open-source toolkit today.
1
/plugin marketplace add exmergo/exmergo-agent-plugins2
/plugin install dex@exmergoRun each command in Claude Code, one at a time.