Generate a Studio artifact

September 29, 2026

Table of contents

  1. Request Headers
  2. Request Body
  3. Responses
  4. Model
  5. Examples
  6. Try It

Generate any Gemini Notebook Studio artifact from a notebook’s sources: an audio overview, a video overview, a slide deck, a report, an interactive report, a data table, a quiz, flashcards, an infographic or a mind map. Every generation runs as a job. Poll it with GET /jobs/jobid, or pass replyUrl to receive the finished record by webhook.

There are two ways to call it:

  • On a notebook — pass the notebook id from POST /notebooks. Its sources must be added first with POST /sources or POST /sources/upload.
  • One-shot — pass urls and/or text instead of notebook. The API creates a notebook for you on one of your accounts, adds the sources, waits for them to be ready and starts the generation. See One-shot requests.

Text results (reports, tables, quiz and flashcard JSON, mind maps) come back inside the job record. Media files (m4a, mp4, pdf, pptx, png and each slide image) are streamed from Google through GET /artifacts/download, and the job result links to it. Those links need your API token, just like every other call.

A finished artifact also stays in its notebook, where GET /artifacts lists it and GET /artifacts/artifact returns it again at any time.

Options by type

Each type accepts only the options in its row. Passing any other option returns 400 Parameter <name> does not apply to type <type>, and a value outside the listed set returns 400 with the valid values. Omitted options use the default shown.

type format length style Other options language instructions
audio deep_dive (default)
brief
critique
debate		 short
default (default)
long		 ❌ — ✅ ✅
video explainer (default)
brief
cinematic
short		 ❌ auto (default)
classic
whiteboard
heritage
paper_craft
watercolor
anime
retro_print
kawaii
custom		 stylePrompt
(with style custom only)		 ✅
(not cinematic)		 ✅
(not cinematic)
report briefing (default)
study_guide
blog_post
custom		 ❌ ❌ — ✅ ✅
(required with custom)
interactive_report ❌ ❌ ❌ — ✅ ❌
table ❌ ❌ ❌ — ✅ ✅
quiz ❌ ❌ ❌ quantity: fewer, standard (default), more
difficulty: easy, medium (default), hard		 ❌ ✅
flashcards ❌ ❌ ❌ quantity: fewer, standard (default), more
difficulty: easy, medium (default), hard		 ❌ ✅
infographic ❌ ❌ auto (default)
sketch_note
professional
bento_grid
editorial
instructional
bricks
clay
anime
kawaii
scientific		 orientation: landscape (default), portrait, square
detail: concise, standard (default), detailed		 ✅ ✅
slides detailed (default)
presenter		 short
default (default)
long		 ❌ — ✅ ✅
mindmap ❌ ❌ ❌ — ❌ ✅

Video rules:

  • cinematic takes no style, stylePrompt, instructions or language.
  • short has a fixed style, so it takes no style or stylePrompt. It does take language and instructions.
  • stylePrompt is required with style: "custom" and rejected with any other style.

What you get back

The job’s result carries the artifact title plus the fields below. The typical times were measured on a Google AI Pro account with three jobs running at once (September 2026). Google’s own cost estimate, as a percentage of the 5-hour usage window, is shown for the Free, Pro and Ultra plans. Both change over time, so read the live estimates from GET /accounts/email.

type result fields Download formats Typical time Estimate, Free / Pro / Ultra
audio duration (seconds), files m4a ~5 min (short debate) 44.1 % / 11.03 % / 0.551 %
video explainer / brief duration (seconds), files mp4 10–12 min 43.7 % / 10.92 % / 0.546 %
video short duration (seconds), files mp4 ~13 min 34.9 % / 8.73 % / 0.437 %
video cinematic duration (seconds), files mp4 not measured 308 % (not allowed on Free) / 77.02 % / 3.85 %
report text (Markdown) — ~100 s 4.3 % / 1.08 % / 0.0542 %
interactive_report text (the report’s paragraphs as plain text) — ~35 s no separate estimate
table table (rows of cells, the first row is the header) — ~30 s 5.9 % / 1.48 % / 0.0737 %
quiz content (JSON: quiz[] with question, answerOptions[], hint) — ~90 s 2.3 % / 0.57 % / 0.0288 %
flashcards content (JSON: flashcards[]) — ~20 s 1.9 % / 0.47 % / 0.0233 %
infographic files (with width / height), text png ~85 s 10.7 % / 2.68 % / 0.134 %
slides files, slides[] (image, caption, text) pdf, pptx, and slide with index for each slide image (PNG) ~7 min 54.5 % / 13.63 % / 0.682 %
mindmap content (JSON tree of name / children) — ~35 s 1.5 % / 0.38 % / 0.0192 %

Google refuses a job once its estimate is larger than what is left of the usage window, even before the window is full. While a job runs, Google holds its full estimate against the window, then settles to the actual cost, which is often much lower. How a refusal comes back depends on the request (see the 429 tab under Responses):

  • On a notebook, Google refuses before the job exists. The call answers 429 with error, retryAt and window at the top level. POST /artifacts/retry and POST /artifacts/revise answer the same way.
  • One-shot, the generation starts inside the job, once its sources are ready. The job fails with error.code: "quota" and error.retryAt, without window. In sync mode, when this happens within the 90-second wait, the call answers 429 with that failed job record. In async mode the call has already answered 201, and GET /jobs/jobid shows the failure.

retryAt and window are present only when Google’s refusal names the window.

Sync, async and webhooks

  • mode: "sync" (the default) waits up to 90 seconds for the result. A job that finishes in time returns 200 with its result. A job still running at 90 seconds returns 202 with the job record, and you keep polling GET /jobs/jobid. If your connection drops during the wait, the job carries on and can still be polled.
  • mode: "async" returns 201 with the job record at once.
  • The API advances running jobs every ~15 seconds. A job still running after 40 minutes is failed with error.code: "timeout". Google may still finish the artifact later, and it then shows up in GET /artifacts.
  • replyUrl receives one POST of the final job record when the job completes or fails.

Each account runs at most maxJobs jobs at a time (default 3, range 1 to 10, set with POST /accounts). GET /jobs shows what is running on each account. DELETE /jobs/jobid frees a slot at once, although Google may still finish the artifact in the notebook.

One-shot requests

Leave out notebook and pass urls and/or text. The API then:

  1. Picks one of your accounts that is healthy, has a free maxJobs slot and whose last known usage does not block this type, or uses the account named by email.
  2. Creates a notebook titled title (default useapi <type>) and adds the sources, with YouTube links added as YouTube sources.
  3. Waits up to 5 minutes for the sources to be ready, then starts the generation. Sources Google could not process are left out and listed in the job’s warnings.

The job’s notebook field names the new notebook, so you can reuse it for more artifacts or chat. Set deleteNotebook: true to have it deleted once the job finishes. That is allowed only for the text-only types (report, interactive_report, table, quiz, flashcards, mindmap), because media files are streamed from their notebook and would go with it. A one-shot notebook is also deleted when the job fails before Google created the artifact, since it holds nothing you can use.

https://api.useapi.net/v1/gemini-notebook/artifacts

Request Headers 
Authorization: Bearer {API token}
Content-Type: application/json
Request Body

On a notebook:

{
  "notebook": "user:[email protected]:d02c903b-17f8-4241-a263-2be9e7359d55",
  "type": "audio",
  "format": "debate",
  "length": "short",
  "language": "es",
  "instructions": "Focus on the risks the crew took during the landing.",
  "mode": "async",
  "replyUrl": "https://your-domain.com/webhook",
  "replyRef": "apollo-debate-1"
}

One-shot:

{
  "type": "quiz",
  "urls": ["https://en.wikipedia.org/wiki/Apollo_11"],
  "title": "Apollo 11 quiz",
  "quantity": "more",
  "difficulty": "hard",
  "deleteNotebook": true
}
  • type is required, the Studio artifact to generate.
    Supported values: audio, video, report, interactive_report, table, quiz, flashcards, infographic, slides, mindmap.
  • notebook is optional, the notebook id from POST /notebooks or GET /notebooks. The id names its account, so the job runs there. Omit it for a one-shot request.
  • sources is optional, with notebook only. An array of 1 to 300 source ids from that notebook to generate from.
    Default: every source in the notebook whose status is ready. The call returns 409 if the notebook has no sources, or none are ready yet.
  • urls is optional, one-shot only. An array of 1 to 50 http(s) URLs. Links on youtube.com or youtu.be are added as YouTube sources, anything else as a web page.
  • text is optional, one-shot only. Pasted text added as one source, up to 500,000 characters.
    A one-shot request needs urls, text, or both.
  • title is optional, one-shot only. The new notebook’s title, also used as the pasted text source’s title. Up to 200 characters.
    Default: useapi <type> for the notebook, Pasted text for the text source.
  • email is optional, one-shot only. The account to run on. When omitted, the API picks a healthy account with a free slot whose usage allows this type.
  • deleteNotebook is optional, one-shot only. true deletes the one-shot notebook once the job completes or fails. Only for report, interactive_report, table, quiz, flashcards and mindmap.
    Default: false.
  • format is optional, see Options by type.
    audio: deep_dive (default), brief, critique, debate.
    video: explainer (default), brief, cinematic, short.
    report: briefing (default), study_guide, blog_post, custom. The first three use Google’s own preset prompts, and any instructions you send are added after the preset’s prompt. custom uses your instructions as the whole prompt.
    slides: detailed (default), presenter.
  • length is optional.
    audio: short, default (default), long.
    slides: short, default (default), long.
  • style is optional.
    video (explainer and brief only): auto (default), classic, whiteboard, heritage, paper_craft, watercolor, anime, retro_print, kawaii, custom.
    infographic: auto (default), sketch_note, professional, bento_grid, editorial, instructional, bricks, clay, anime, kawaii, scientific.
  • stylePrompt is optional, video only. Describes your own visual style, up to 2,000 characters. Required with style: "custom", and rejected otherwise.
  • quantity is optional, quiz and flashcards only.
    Supported values: fewer, standard (default), more.
  • difficulty is optional, quiz and flashcards only.
    Supported values: easy, medium (default), hard.
  • orientation is optional, infographic only.
    Supported values: landscape (default), portrait, square.
  • detail is optional, infographic only.
    Supported values: concise, standard (default), detailed.
  • language is optional, the output language as a language code such as en, es, pt_BR or zh_Hans. Accepted by audio, video (not cinematic), report, interactive_report, table, infographic and slides.
    Default: en.
  • instructions is optional, what to focus on or how to shape the result, up to 10,000 characters. Accepted by audio, video (not cinematic), table, quiz, flashcards, infographic, slides and mindmap. For report with format: "custom" it is the whole prompt and is required. With a preset report format it is added after the preset’s own prompt.
  • mode is optional, sync (default) or async. See Sync, async and webhooks.
  • replyUrl is optional, a public http(s) URL that receives one POST of the job record when the job completes or fails. See GET /jobs/jobid for the record’s shape.
  • replyRef is optional, your own reference echoed back in the job record, up to 1024 characters.
Responses

  • 200 OK

    Sync mode, finished within 90 seconds. The result shape depends on the type, see What you get back.

    {
  "jobid": "job:0ff1aa90-5244-4900-9e7d-e0931ccce82e-user:[email protected]:gemini_notebook",
  "email": "[email protected]",
  "type": "infographic",
  "status": "completed",
  "notebook": "user:[email protected]:d02c903b-17f8-4241-a263-2be9e7359d55",
  "artifact": "user:[email protected]:d02c903b-17f8-4241-a263-2be9e7359d55:9213e77e-9b71-4f84-8f71-1bd130138a01",
  "created_at": "2026-09-27T23:31:07.016Z",
  "request": {
    "notebook": "user:[email protected]:d02c903b-17f8-4241-a263-2be9e7359d55",
    "type": "infographic",
    "style": "sketch_note",
    "orientation": "portrait"
  },
  "completed_at": "2026-09-27T23:32:42.962Z",
  "result": {
    "title": "Apollo 11 Moon Landing Infographic",
    "files": [
      {
        "format": "png",
        "mimeType": "image/png",
        "url": "https://api.useapi.net/v1/gemini-notebook/artifacts/download?artifact=user%3A12345-user%40example.com-artifact%3Ad02c903b-17f8-4241-a263-2be9e7359d55%3A9213e77e-9b71-4f84-8f71-1bd130138a01&format=png",
        "width": 1536,
        "height": 2752
      }
    ],
    "text": "APOLLO 11: ONE GIANT LEAP FOR MANKIND\n\n[An illustrated infographic detailing the key stages of the 1969 moon landing mission...]\n\nMISSION PARAMETERS AT A GLANCE\n\n- Spacecraft: CSM-107 Columbia / LM-5 Eagle\n..."
  }
}

    A one-shot quiz, with its JSON content:

    {
  "jobid": "job:4855bb5a-6292-4419-9fb8-1d42b81e3a78-user:[email protected]:gemini_notebook",
  "email": "[email protected]",
  "type": "quiz",
  "status": "completed",
  "notebook": "user:[email protected]:38790668-c173-4f95-9954-bb482739b582",
  "created_at": "2026-09-28T00:08:21.063Z",
  "request": {
    "type": "quiz",
    "urls": ["https://en.wikipedia.org/wiki/Apollo_11"],
    "title": "useapi one-shot",
    "deleteNotebook": true
  },
  "updated_at": "2026-09-28T00:08:25.396Z",
  "artifact": "user:[email protected]:38790668-c173-4f95-9954-bb482739b582:555b18c1-237c-4014-ae8f-b5b931d2bde7",
  "completed_at": "2026-09-28T00:09:29.141Z",
  "result": {
    "title": "Apollo Quiz",
    "content": {
      "quiz": [
        {
          "question": "Why did President John F. Kennedy select a crewed lunar landing mission as the target for the United States in 1961...?",
          "answerOptions": [
            {
              "text": "The Soviet Union had launch vehicles with higher lift capacity, so Kennedy chose a challenge requiring new rocket technology...",
              "rationale": "Selecting a goal beyond existing rocket capabilities ensured that the Soviet Union's existing advantage...",
              "isCorrect": true
            },
            {
              "text": "The United States already possessed Saturn V rockets capable of deep space flight...",
              "rationale": "The Saturn V rocket was still in early development phases...",
              "isCorrect": false
            }
          ],
          "hint": "Consider how initial Soviet advantages in rocket payload capacity influenced American strategic goal setting."
        }
      ]
    }
  }
}

  • 201 Created

    Async mode. The job is created and returned at once. Poll GET /jobs/jobid until status is completed or failed. A one-shot job starts as pending without an artifact, which appears once Google has created it.

    {
  "jobid": "job:12597f9a-010c-4758-8e9a-b5cc17b0ccdb-user:[email protected]:gemini_notebook",
  "email": "[email protected]",
  "type": "audio",
  "status": "processing",
  "notebook": "user:[email protected]:d02c903b-17f8-4241-a263-2be9e7359d55",
  "artifact": "user:[email protected]:d02c903b-17f8-4241-a263-2be9e7359d55:0c2ba8dc-47dd-474f-a19c-c06896e269fd",
  "created_at": "2026-09-27T23:38:15.692Z",
  "request": {
    "notebook": "user:[email protected]:d02c903b-17f8-4241-a263-2be9e7359d55",
    "type": "audio",
    "format": "debate",
    "length": "short",
    "language": "es",
    "instructions": "Focus on the risks the crew took during the landing."
  },
  "replyUrl": "https://your-domain.com/webhook",
  "replyRef": "apollo-debate-1"
}

  • 202 Accepted

    Sync mode, still running after 90 seconds. Poll GET /jobs/jobid for the result. The job record is the same as the 201 one.

    {
  "jobid": "job:7e575d08-5298-4a75-bd51-03de14ff3e31-user:[email protected]:gemini_notebook",
  "email": "[email protected]",
  "type": "video",
  "status": "processing",
  "notebook": "user:[email protected]:d02c903b-17f8-4241-a263-2be9e7359d55",
  "artifact": "user:[email protected]:d02c903b-17f8-4241-a263-2be9e7359d55:5da139a1-f97b-443e-9ef5-629a9e91dc37",
  "created_at": "2026-09-27T23:38:38.606Z",
  "request": {
    "notebook": "user:[email protected]:d02c903b-17f8-4241-a263-2be9e7359d55",
    "type": "video",
    "format": "brief",
    "style": "custom",
    "stylePrompt": "1960s newsreel, black and white"
  }
}

  • 400 Bad Request

    A missing or invalid parameter, or an option that does not fit the type.

    {
  "error": "Parameter style does not apply to type slides",
  "code": 400
}

    {
  "error": "deleteNotebook is only for report, interactive_report, table, quiz, flashcards, mindmap: a video's files are served from its notebook",
  "code": 400
}

  • 401 Unauthorized

    Invalid API token.

    {
  "error": "useapi.net ⁝ Unauthorized",
  "code": 401
}

  • 403 Forbidden

    The notebook id belongs to a different API token, or Google does not allow this on the account’s plan.

    {
  "error": "user:[email protected]:d02c903b-17f8-4241-a263-2be9e7359d55 does not belong to this API token",
  "code": 403
}

  • 404 Not Found

    The notebook no longer exists at Google, or the account it names (or the email you passed) is not configured.

    {
  "error": "Not found on account [email protected] (Google rLM1Ne grpc 5)",
  "code": 404
}

  • 409 Conflict

    The notebook has no sources, or none of them are ready yet. Check each source’s status with GET /notebooks/notebook.

    {
  "error": "The notebook's sources are still processing, retry shortly",
  "code": 409
}

  • 422 Unprocessable Content

    Sync mode, the job failed within 90 seconds: Google reported the generation as failed (generation_failed), or no source of a one-shot request could be used (sources). A failed artifact can be re-run with POST /artifacts/retry.

    {
  "jobid": "job:3b1f6c2e-8d47-4a95-b0e2-7c9d51a4e8f3-user:[email protected]:gemini_notebook",
  "email": "[email protected]",
  "type": "quiz",
  "status": "failed",
  "notebook": "user:[email protected]:d02c903b-17f8-4241-a263-2be9e7359d55",
  "artifact": "user:[email protected]:d02c903b-17f8-4241-a263-2be9e7359d55:5e2a9c71-4b3d-4f08-9a6e-d18c73b2f045",
  "created_at": "2026-09-27T23:26:40.393Z",
  "request": {
    "notebook": "user:[email protected]:d02c903b-17f8-4241-a263-2be9e7359d55",
    "type": "quiz"
  },
  "completed_at": "2026-09-27T23:27:25.114Z",
  "error": {
    "code": "generation_failed",
    "message": "Google reported the generation as failed. POST /artifacts/retry can re-run it in place"
  }
}

  • 429 Too Many Requests

    Returned in four cases.

    The account already runs maxJobs jobs. Retry when one finishes, or raise maxJobs with POST /accounts:

    {
  "error": "Account [email protected] is busy: 3/3 jobs running (maxJobs)",
  "code": 429
}

    One-shot without email, and every account is busy, or blocked by its last known usage:

    {
  "error": "All 2 accounts are at their maxJobs limit, retry shortly",
  "code": 429
}

    {
  "error": "Google's usage limit blocks audio on every available account, check GET /accounts/{email} for reset times",
  "code": 429
}

    On a notebook, Google refused the job because not enough of its usage window is left. No job was created. retryAt is when the window resets and window is 5h or weekly, both present only when Google’s refusal names the window. GET /accounts/email shows what is left and what each type needs:

    {
  "error": "Not enough of Google's 5h window is left on account [email protected] for this job; it resets at 2026-09-28T04:25:40.000Z. GET /accounts/[email protected] shows what each job needs",
  "code": 429,
  "retryAt": "2026-09-28T04:25:40.000Z",
  "window": "5h"
}

    One-shot in sync mode, Google refused the generation once the job’s sources were ready. The body is the failed job record, with error.code: "quota" and error.retryAt (no window). The one-shot notebook is deleted, since Google created no artifact in it. In async mode the call answers 201 instead, and the job fails the same way:

    {
  "jobid": "job:9c4e2f1a-7b3d-4e85-a6c2-51d8f0b3e7a4-user:[email protected]:gemini_notebook",
  "email": "[email protected]",
  "type": "audio",
  "status": "failed",
  "notebook": "user:[email protected]:6a1f3c9e-2d48-4b7a-9e05-c3f81d7a2b64",
  "created_at": "2026-09-28T01:12:04.518Z",
  "request": {
    "type": "audio",
    "urls": ["https://en.wikipedia.org/wiki/Apollo_11"]
  },
  "completed_at": "2026-09-28T01:12:31.207Z",
  "error": {
    "code": "quota",
    "message": "Not enough of Google's 5h window is left on account [email protected] for this job; it resets at 2026-09-28T04:25:40.000Z. GET /accounts/[email protected] shows what each job needs",
    "retryAt": "2026-09-28T04:25:40.000Z"
  }
}

  • 502 Bad Gateway

    Google answered with an unexpected error, or could not be reached. In sync mode a job that failed for such a reason also returns 502 with the failed job record.

    {
  "error": "Could not reach Gemini Notebook (HTTP 500), please retry",
  "code": 502
}

  • 503 Service Unavailable

    A temporary condition, retry in a few minutes.

    {
  "error": "GOOGLE_BOT_CHECK: Google is challenging our network right now, please retry in a few minutes",
  "code": 503
}

    {
  "error": "Async mode is unavailable right now, use mode sync",
  "code": 503
}

    The job could not be added to our job schedule. It is failed at once with error.code: "scheduler". Google may already have started the generation, since only a one-shot request creates the artifact inside the job. Check GET /artifacts before you submit it again.

    {
  "error": "Unable to lock the job schedule of user 12345, please retry",
  "code": 503
}

    {
  "error": "Failed to create the job schedule (QStash HTTP 500): …",
  "code": 503
}

  • 596 Account Error

    Google reported the account as signed out. While we re-check it, the answer is the first message below. Retry in about a minute. If it stays signed out, the account is paused, you receive an email, and the answer becomes the second message until you re-add it via Setup Gemini Notebook.

    {
  "error": "Google reported this account as signed out. We are re-checking it — please retry in about a minute. If it stays signed out, you will receive a re-add email.",
  "code": 596
}

    {
  "error": "Account [email protected]: Google signed this account out. Re-add it at https://useapi.net/docs/start-here/setup-gemini-notebook",
  "code": 596
}

    In sync mode, a job whose account is confirmed signed out while it runs answers 596 with the failed job record (error.code: "account").

    A one-shot request without email returns 596 when none of your accounts is healthy:

    {
  "error": "No healthy Gemini Notebook accounts available. Check GET /accounts",
  "code": 596
}
Model

The job record. GET /jobs/jobid and the replyUrl webhook return the same shape.

In sync mode a failed job answers with an HTTP status that matches error.code: quota → 429, account → 596, not_found → 404, sources and generation_failed → 422, cancelled → 409, timeout → 504, anything else → 502.

{ // TypeScript, all fields are optional
  jobid: string                  // job:<uuid>-user:<id>-<email>-bot:gemini_notebook
  email: string                  // the account the job runs on
  type: 'audio' | 'video' | 'report' | 'interactive_report' | 'table' | 'quiz' | 'flashcards' | 'infographic' | 'slides' | 'mindmap' | 'research'
  status: 'pending' | 'processing' | 'completed' | 'failed'
  notebook: string               // notebook id; for a one-shot, the notebook created for it
  artifact?: string              // artifact id, once Google has created it (never on a research job)
  created_at: string             // ISO 8601, when the job was accepted
  updated_at?: string            // one-shot only: when its artifact was created
  completed_at?: string          // ISO 8601, when the job completed or failed
  request: Record<string, unknown>   // your request body echoed back (ids as strings), without mode, replyUrl and replyRef
  replyUrl?: string
  replyRef?: string
  warnings?: string[]            // one-shot only: sources Google could not process and that were left out
  error?: {                      // status 'failed'
    code: 'quota' | 'account' | 'not_found' | 'sources' | 'generation_failed' | 'timeout' | 'google_error' | 'scheduler' | 'cancelled'
    message: string
    retryAt?: string             // code 'quota', when Google names the window: ISO 8601 time it resets
  }
  result?: {                     // status 'completed', a Studio artifact
    title: string
    duration?: number            // audio, video: seconds
    files?: Array<{              // audio, video, infographic, slides
      format: 'm4a' | 'mp4' | 'png' | 'pdf' | 'pptx'
      mimeType: string
      url: string                // GET /artifacts/download link, needs your API token
      width?: number             // infographic
      height?: number
    }>
    slides?: Array<{             // slides
      image: string              // GET /artifacts/download link (format=slide&index=N)
      caption: string | null
      text: string | null
    }>
    text?: string                // report (Markdown), interactive_report, infographic
    table?: string[][]           // table: rows of cells, header row first
    content?: unknown            // quiz, flashcards, mindmap: Google's JSON
  } | {                          // status 'completed', type 'research' (POST /research)
    mode: 'fast' | 'deep'
    query: string
    summary?: string             // fast: Google's one-line summary of what it found
    report?: {                   // deep: the Deep Research report
      title: string
      markdown: string
    }
    noResults?: true             // the run found nothing (sources is [])
    sources: Array<{
      url: string
      title: string
      description: string | null
      cited: boolean             // deep: the report cites it. fast: always true
    }>
  }
}
Examples 
  • curl -X POST "https://api.useapi.net/v1/gemini-notebook/artifacts" \
   -H "Content-Type: application/json" \
   -H "Authorization: Bearer …" \
   -d '{
     "notebook": "user:[email protected]:d02c903b-17f8-4241-a263-2be9e7359d55",
     "type": "slides",
     "format": "presenter",
     "length": "short",
     "mode": "async"
   }'

  • const token = "API token";
const apiUrl = "https://api.useapi.net/v1/gemini-notebook";
const headers = { "Content-Type": "application/json", "Authorization": `Bearer ${token}` };

const response = await fetch(`${apiUrl}/artifacts`, {
  method: "POST",
  headers,
  body: JSON.stringify({
    type: "audio",
    urls: ["https://en.wikipedia.org/wiki/Apollo_11"],
    format: "brief",
    mode: "async"
  })
});
let job = await response.json();
console.log("submitted", response.status, job);

while (job.status === "pending" || job.status === "processing") {
  await new Promise(r => setTimeout(r, 15000));
  job = await (await fetch(`${apiUrl}/jobs/${encodeURIComponent(job.jobid)}`, { headers })).json();
  console.log(job.status);
}
console.log("result", job.result ?? job.error);

  • import time
import requests
from urllib.parse import quote

token = "API token"
apiUrl = "https://api.useapi.net/v1/gemini-notebook"
headers = {"Content-Type": "application/json", "Authorization": f"Bearer {token}"}

data = {
    "type": "audio",
    "urls": ["https://en.wikipedia.org/wiki/Apollo_11"],
    "format": "brief",
    "mode": "async"
}
response = requests.post(f"{apiUrl}/artifacts", headers=headers, json=data)
job = response.json()
print("submitted", response.status_code, job)

while job.get("status") in ("pending", "processing"):
    time.sleep(15)
    job = requests.get(f"{apiUrl}/jobs/{quote(job['jobid'], safe='')}", headers=headers).json()
    print(job["status"])

print("result", job.get("result") or job.get("error"))
Try It