> ## Documentation Index
> Fetch the complete documentation index at: https://agent-compass.mintlify.site/llms.txt
> Use this file to discover all available pages before exploring further.

# GDPval AC

> Economically valuable task benchmark adapted for AgentCompass.

GDPval AC evaluates economically valuable real-world tasks and compares generated deliverables against a fixed baseline when judge assets are staged.

## Runtime Status

| Field               | Value                                          |
| ------------------- | ---------------------------------------------- |
| Benchmark id        | `gdpval_ac`                                    |
| Tags                | `Productivity`, `Long-Horizon`, `Judge-Scored` |
| Execution type      | remote environment                             |
| Typical harness     | compatible productivity or coding harness      |
| Typical environment | recipe-backed remote provider                  |
| Current status      | registered in the direct runtime               |

## Run Pattern

```bash theme={null}
agentcompass run \
  gdpval_ac \
  <compatible-harness> \
  your-model \
  --env <env-provider> \
  --benchmark-params '{"sample_ids":["<task-id>"]}' \
  --model-base-url "$MODEL_BASE_URL" \
  --model-api-key "$MODEL_API_KEY"
```

## Common Parameters

| Parameter     | Meaning                                            |
| ------------- | -------------------------------------------------- |
| `sample_ids`  | Run selected task ids.                             |
| `split`       | Dataset split.                                     |
| `judge_model` | Judge model spec when pairwise judging is enabled. |

## Outputs

Per-task details are written to `results/gdpval_ac/<model>/<run>/details/`. Aggregate metrics are written to `summary.md`.