> ## Documentation Index
> Fetch the complete documentation index at: https://docs.agentset.ai/llms.txt
> Use this file to discover all available pages before exploring further.

# File Uploads

> Upload files to Agentset

Upload documents to Agentset for processing. You can either provide a URL to a publicly accessible file, or upload files directly from your application.

## Supported file types

Agentset supports the following file formats for ingestion.

| Type         | Extensions                                        |
| :----------- | :------------------------------------------------ |
| PDF          | `.pdf`                                            |
| Spreadsheet  | `.csv`, `.xls`, `.xlsx`, `.ods`                   |
| Word         | `.doc`, `.docx`, `.odt`                           |
| Presentation | `.ppt`, `.pptx`, `.odp`                           |
| HTML         | `.html`, `.htm`                                   |
| EPUB         | `.epub`                                           |
| Image        | `.png`, `.jpg`, `.jpeg`, `.webp`, `.gif`, `.tiff` |
| Outlook      | `.msg`                                            |
| Plain text   | `.txt`                                            |
| Markdown     | `.md`                                             |
| JSON         | `.json`, `.jsonl`                                 |
| XML          | `.xml`                                            |
| RSS          | `.rss`, `.atom`                                   |

## Upload from URL

Provide a URL to a file and Agentset will fetch and process it.

<CodeGroup>
  ```typescript TypeScript theme={null}
  import { Agentset } from "agentset";

  const agentset = new Agentset({
    apiKey: process.env.AGENTSET_API_KEY,
  });

  const ns = agentset.namespace("YOUR_NAMESPACE_ID");

  const job = await ns.ingestion.create({
    payload: {
      type: "FILE",
      fileUrl: "https://example.com/document.pdf",
    },
  });

  console.log(`Upload started: ${job.id}`);
  ```

  ```python Python theme={null}
  import os
  from agentset import Agentset

  client = Agentset(
      namespace_id="YOUR_NAMESPACE_ID",
      token=os.environ["AGENTSET_API_KEY"],
  )

  job = client.ingest_jobs.create(
      payload={
          "type": "FILE",
          "fileUrl": "https://example.com/document.pdf",
      }
  )

  print(f"Upload started: {job.data.id}")
  ```
</CodeGroup>

## Direct upload

Upload files directly from your application using presigned URLs. This is useful when files aren't publicly accessible.

<CodeGroup>
  ```typescript TypeScript theme={null}
  import fs from "fs";

  const upload = await ns.uploads.upload({
    file: fs.createReadStream("./document.pdf"),
    contentType: "application/pdf",
  });

  // Create an ingest job for the uploaded file
  const job = await ns.ingestion.create({
    payload: {
      type: "MANAGED_FILE",
      key: upload.key,
      fileName: "document.pdf",
    },
  });

  console.log(`Upload started: ${job.id}`);
  ```

  ```python Python theme={null}
  import os

  with open("./document.pdf", "rb") as f:
      file = f.read()

  # Get a presigned upload URL
  upload = client.uploads.create(
      file_name="document.pdf",
      file_size=len(file),
      content_type="application/pdf",
  )

  # Upload the file
  import requests
  requests.put(
      upload.data.url,
      data=file,
      headers={"Content-Type": "application/pdf"},
  )

  # Create an ingest job for the uploaded file
  job = client.ingest_jobs.create(
      payload={
          "type": "MANAGED_FILE",
          "key": upload.data.key,
          "fileName": "document.pdf",
      }
  )

  print(f"Upload started: {job.data.id}")
  ```
</CodeGroup>

## With metadata

Attach metadata to your files for [filtering](/search-and-retrieval/filtering) during search.

<CodeGroup>
  ```typescript TypeScript theme={null}
  const job = await ns.ingestion.create({
    name: "Q4 2024 Report",
    payload: {
      type: "FILE",
      fileUrl: "https://example.com/quarterly-report.pdf",
    },
    config: {
      metadata: {
        quarter: "Q4",
        year: "2024",
        type: "financial",
      },
    },
  });
  ```

  ```python Python theme={null}
  job = client.ingest_jobs.create(
      name="Q4 2024 Report",
      payload={
          "type": "FILE",
          "fileUrl": "https://example.com/quarterly-report.pdf",
      },
      config={
          "metadata": {
              "quarter": "Q4",
              "year": "2024",
              "type": "financial",
          },
      },
  )
  ```
</CodeGroup>

## File size limit

The maximum file size is 200 MB.

<Info>
  File uploads are processed asynchronously. Learn how to [check upload status](/data-ingestion/upload-status).
</Info>

## Next steps

* [API Reference](/api-reference/endpoint/ingest-jobs/create) — File upload parameters and options
* [Document Metadata](/data-ingestion/document-metadata) — Learn more about metadata filtering
* [Chunking Settings](/data-ingestion/chunking-settings) — Configure how files are split into chunks
* [Multimodal Input](/data-ingestion/multimodal-input) — Process images and other media
* [Search](/search-and-retrieval/search) — Query your uploaded content