Studies
The studies endpoints return GPTfake’s research studies — longitudinal analyses of how model behavior drifts over time, and focused studies on specific topics or policy areas. These map to the work published on our research hub, including longitudinal studies and policy analysis.
List studies
GET /research/studiesQuery parameters:
| Param | Values | Default |
|---|---|---|
type | longitudinal, policy, topical | all |
model | any model id | all |
year | e.g. 2024 | all |
curl "https://api.gptfake.com/v1/research/studies?type=longitudinal" \
-H "Authorization: Bearer YOUR_API_KEY"Response:
{
"studies": [
{
"id": "study_drift_2024",
"title": "Policy Drift Across Major LLMs, 2023–2024",
"type": "longitudinal",
"models": ["chatgpt", "claude", "gemini"],
"published": "2024-10-15",
"dataset_id": "ds_drift_2024",
"url": "https://gptfake.com/research/longitudinal-studies"
}
]
}Get a single study
GET /research/studies/{study_id}curl https://api.gptfake.com/v1/research/studies/study_drift_2024 \
-H "Authorization: Bearer YOUR_API_KEY"The response includes the study abstract, methodology link, the models analyzed, and the dataset_id you can export.
Every study links to the methodology it followed and to its backing dataset, so results are reproducible. This is core to our E-E-A-T and watchdog independence.
Next steps
- Turn a study’s
dataset_idinto a download via Data export. - Cite the formal write-up — see Publications.
GPTfake