---
title: REST API | Developer Documentation
description: Guide on how to use the LlamaExtract REST API for programmatic data extraction, including creating agents, uploading documents, and running extraction jobs.
---

## Quickstart

### Create a new agent

```
curl -X 'POST' \
  'https://api.cloud.llamaindex.ai/api/v1/extraction/extraction-agents' \
  -H 'accept: application/json' \
  -H "Authorization: Bearer $LLAMA_CLOUD_API_KEY" \
  -H 'Content-Type: application/json' \
  -d '{
    "name": "resume_parser",
    "data_schema": {
        "type": "object",
        "properties": {
            "name": {
                "type": "string",
                "description": "Candidate name"
            },
            "experience": {
                "type": "array",
                "description": "Work history",
                "items": {
                    "type": "object",
                    "properties": {
                        "company": {
                            "type": "string",
                            "description": "Company name"
                        },
                        "title": {
                            "type": "string",
                            "description": "Job title"
                        },
                        "start_date": {
                            "anyOf": [
                                {
                                    "type": "string"
                                },
                                {
                                    "type": "null"
                                }
                            ],
                            "description": "Start date of employment"
                        },
                        "end_date": {
                            "anyOf": [
                                {
                                    "type": "string"
                                },
                                {
                                    "type": "null"
                                }
                            ],
                            "description": "End date of employment"
                        }
                    }
                }
            }
        }
    },
    "config": {
        "extraction_target": "PER_DOC",
        "extraction_mode": "BALANCED",
    }
}'
```

You can list all your existing agents with the following command. The `project_id` can be obtained from the [web UI](../web_ui).

Also refer to the the [Projects API](https://developers.llamaindex.ai/cloud-api-reference/category/projects) for programmatically creating/editing projects.

```
curl -X 'GET' \
  'https://api.cloud.llamaindex.ai/api/v1/extraction/extraction-agents?project_id={project_id}' \
  -H 'accept: application/json' \
  -H "Authorization: Bearer $LLAMA_CLOUD_API_KEY"
```

You can also fetch a specific `agent_id` by name:

```
curl -X 'GET' \
  'https://api.cloud.llamaindex.ai/api/v1/extraction/extraction-agents/by-name/{agent_name}' \
  -H 'accept: application/json' \
  -H "Authorization: Bearer $LLAMA_CLOUD_API_KEY"
```

### Upload a document

Upload a document using our [Upload API](https://developers.llamaindex.ai/cloud-api-reference/upload-file-api-v-1-files-post).

```
curl -X 'POST' \
  'https://api.cloud.llamaindex.ai/api/v1/files' \
  -H 'accept: application/json' \
  -H 'Content-Type: multipart/form-data' \
  -H "Authorization: Bearer $LLAMA_CLOUD_API_KEY" \
  -F 'upload_file=@/path/to/your/file.pdf;type=application/pdf'
```

### Run an extraction job

Use the `extraction_agent_id` and `file_id` from the previous steps to run an extraction job.

```
curl -X 'POST' \
  'https://api.cloud.llamaindex.ai/api/v1/extraction/jobs' \
  -H 'accept: application/json' \
  -H 'Content-Type: application/json' \
  -H "Authorization: Bearer $LLAMA_CLOUD_API_KEY" \
  -d '{
  "extraction_agent_id": "{$AGENT_ID}",
  "file_id": "{$FILE_ID}",
}'
```

### Poll for extraction job status

Jobs are processed asynchronously. You can poll for the status of a job using the following endpoint.

```
curl -X 'GET' \
  'https://api.cloud.llamaindex.ai/api/v1/extraction/jobs/{$JOB_ID}' \
  -H 'accept: application/json' \
  -H "Authorization: Bearer $LLAMA_CLOUD_API_KEY"
```

### Get the results of an extraction job

If the job is successfully completed, you will see the status as `SUCCESS`. You can then retrieve the results from the following endpoint.

```
curl -X 'GET' \
  'https://api.cloud.llamaindex.ai/api/v1/extraction/jobs/{$JOB_ID}/result' \
  -H 'accept: application/json' \
  -H "Authorization: Bearer $LLAMA_CLOUD_API_KEY"
```

## Full API documentation

This is just a subset of the available endpoints to help you get started.

You can see all the available endpoints in our [full API documentation](https://developers.llamaindex.ai/cloud-api-reference/category/llama-extract).