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

# Experiment Fetch Limited to 1000 Events - Export All

export const plans_0 = "Any"

export const deployments_0 = "Any"

export const data_plane_version_0 = undefined

export const use_case_0 = undefined

<Note>
  **Applies to:**

  * Plan - {plans_0}
  * Deployment - {deployments_0}
  * {data_plane_version_0}
  * {use_case_0}
</Note>

## Summary

**Goal:** Export complete experiment data beyond the 1000 event limit using BTQL with cursor pagination.

**Features:** BTQL endpoint, cursor-based pagination via x-bt-cursor header, native query format.

## Configuration Steps

### Step 1: Understand the limitation

The `/v1/experiment/{id}/fetch` endpoint has a hard limit of 1000 events per request and does not support cursor pagination.

### Step 2: Use BTQL endpoint for full export

The `/btql` endpoint supports cursor-based pagination to retrieve all experiment events.

```python theme={"theme":{"light":"github-light","dark":"github-dark-dimmed"}}
import requests
import json

def fetch_all_experiment_events(experiment_id, api_key):
    headers = {
        "Authorization": f"Bearer {api_key}",
        "Content-Type": "application/json"
    }
    all_events = []
    cursor = None

    while True:
        query = {
            "query": {
                "from": {
                    "op": "function",
                    "name": {"op": "ident", "name": ["experiment"]},
                    "args": [{"op": "literal", "value": experiment_id}]
                },
                "select": [{"op": "star"}],
                "limit": 1000
            },
            "fmt": "jsonl"
        }

        if cursor:
            query["query"]["cursor"] = cursor

        resp = requests.post(
            "https://api.braintrust.dev/btql",
            headers=headers,
            json=query
        )

        # Extract cursor from response header
        cursor = resp.headers.get("x-bt-cursor")

        # Parse JSONL response
        rows = [
            json.loads(line)
            for line in resp.text.strip().split("\n")
            if line.strip()
        ]

        all_events.extend(rows)

        # Stop when no cursor returned
        if not cursor:
            break

    return all_events

# Usage
events = fetch_all_experiment_events(
    "your-experiment-id",
    "your-api-key"
)
# Print example, handle event data however you need from here.
print(f"Total events: {len(events)}")

```

### Step 3: Extract pagination cursor from headers

The pagination cursor is returned in the `x-bt-cursor` response header, not in the response body.

### Step 4: Pass cursor in subsequent requests

Add the cursor to the `query.cursor` field in subsequent requests until `x-bt-cursor` header is null.

## Key Differences

| Feature                | /fetch endpoint   | /btql endpoint     |
| ---------------------- | ----------------- | ------------------ |
| Max events per request | 1000 (hard limit) | 1000 (paginated)   |
| Pagination support     | No                | Yes (cursor-based) |
| Cursor location        | N/A               | x-bt-cursor header |
| Query format           | Simple parameters | Native BTQL format |