List Jobs

Retrieve full job records, including aggregated scores and per-model results, with optional filtering, sorting, and paging. Soft-deleted jobs are excluded.

GET /v1/jobs

Authentication

• Authentication is required. Include a valid df_ API key in the Authorization header, or rely on the machine-local default credential. See Authentication.

---

Query Parameters

All parameters are optional.

Parameter	Type	Default	Description	Required
job_id	number(int)	—	Get for a single job_id	No
status	string	—	Filter by job status (e.g. queued, processing, complete, error).	No
model	string	—	Return only jobs that included this model.	No
file_type	string	—	Filter by video, audio, or image.	No
title	string	—	Case-insensitive substring match on the job title.	No
limit	number(int)	50	Maximum records to return (clamped to 1–500).	No
offset	number(int)	0	Number of records to skip, for paging.	No
sort_by	string	created_at	One of id, status, title, file_type, created_at (date is an alias for created_at).	No
sort_order	string	desc	asc or desc.	No

---

Response

data is an array of full job records.

Field	Type	Description
id	number(int)	The job id.
input_path	string	The submitted file path.
title	string | null	The job label.
mode	string | null	The processing mode.
file_type	string | null	video, audio, or image.
models	string[]	The models selected for the job.
status	string	queued, processing, complete, or error.
error_message	string | null	Failure detail when status is error.
core_job_id	number(int) | null	The underlying core processing id.
created_at / updated_at / completed_at / deleted_at	timestamp | null	Lifecycle timestamps.
video_time / audio_time	number(float) | null	Lane processing time in seconds.
progress	number(float) | null	Completion percentage.
prediction_score	number(float) | null	The final aggregated score across all applicable lanes.
prediction_label	string | null	The verdict: FAKE, REAL, or NONE when no verdict could be produced.
prediction_source	string | null	The scoring profile that scored the job — "default" or "name@version".
video	object	Returned only for video jobs with visual models. Fields: score(float), coverage(float), threshold(float), status(string), error(string), chunk_scores(float[]), optional frame_scores(float[]), optional face_box(float[][]).
audio	object	Returned only for audio/video jobs with audio models. Fields: score(float), coverage(float), threshold(float), status(string), error(string), chunk_scores(float[]), optional frame_scores(float[]).
image	object	Returned only for image jobs. Fields: score(float), coverage(float), threshold(float), status(string), error(string), chunk_scores(float[]).

{
  "status": true,
  "code": 200,
  "message": "Jobs fetched",
  "data": [
    {
      "id": 42,
      "input_path": "/var/lib/deepfaic/uploads/sample.mp4",
      "title": "Q3 onboarding call",
      "mode": "normal",
      "file_type": "video",
      "models": ["facia", "diffusion", "swintf", "audio"],
      "status": "complete",
      "error_message": null,
      "core_job_id": 128,
      "created_at": "2026-06-04T10:15:00Z",
      "updated_at": "2026-06-04T10:16:12Z",
      "completed_at": "2026-06-04T10:16:12Z",
      "deleted_at": null,
      "video_time": 41.2,
      "audio_time": 12.8,
      "progress": 100.0,
      "video": {
        "score": 0.18,
        "coverage": 0.92,
        "threshold": 0.4,
        "status": "complete",
        "error": null,
        "chunk_scores": [0.12, 0.21, 0.17],
        "frame_scores": [0.11, 0.19, 0.22],
        "face_box": [[0.1, 0.2, 0.4, 0.6]]
      },
      "audio": {
        "score": 0.07,
        "coverage": 1.0,
        "threshold": 0.4,
        "status": "complete",
        "error": null,
        "chunk_scores": [0.05, 0.08],
        "frame_scores": [0.04, 0.06, 0.08]
      }
    }
  ]
}

tip

If you only need lifecycle and status fields, use List Jobs (Partial) for a lighter payload.

---

Example Request

curl "http://<edge-host>:9123/v1/jobs?status=complete&file_type=video&limit=20" \
  -H "Authorization: Bearer df_..."