# Concurrency limits

Source: https://docs.scrappey.com/docs/concurrency-limits

> Scrappey gives new accounts up to 200 concurrent threads by default and scales automatically up to 1000, with higher limits available on request.

## How Many Concurrent Threads Can I Use?

Scrappey allows all **new registered users** to start with **up to 200 concurrent threads** by default.  
This is more than enough for the majority of use cases and ensures smooth performance and fair resource distribution.

If your project requires more, Scrappey makes it easy to scale.

## Scaling Beyond 200 Threads

We automatically allow **increases up to 1000 concurrent threads** without manual review, provided:
- Your **overall success rate is above 50%**.
- You are following our general API guidelines.
- You are not abusing endpoints.

## Special Cases / Higher Limits

If you require **more than 1000 concurrent threads**, this is no problem — simply reach out to support with a short explanation of your use case.

We generally approve higher limits quickly for:
- High-throughput data pipelines
- Large-scale monitoring, indexing, and QA workloads
- Long-running automation workflows that require sustained concurrency

There is **no hard upper limit** as long as your use case fits within reasonable resource bounds and success rates remain healthy.

## Quick Summary

| User Type       | Threads Allowed | Requirements          |
|-----------------|-----------------|------------------------|
| New Account     | Up to 200        | None                   |
| Active Account  | Up to 1000       | > 50% success rate      |
| Special Request | > 1000           | Case-by-case approval   |

## Tips for Scaling Smoothly
- **Monitor your success rate** regularly in the dashboard.
- Increase threads gradually to avoid sudden overload.
- If you encounter issues, check error codes in the docs first.

## Example

Concurrency is not a request parameter — it's simply how many requests you have **in flight at the same time**. Each simultaneous request consumes one thread. Fire requests in parallel from your own code and Scrappey processes them concurrently up to your account limit:

```bash
# Fire 5 requests in parallel — each one uses one concurrent thread
for url in \
  "https://example.com/page1" \
  "https://example.com/page2" \
  "https://example.com/page3" \
  "https://example.com/page4" \
  "https://example.com/page5"
do
  curl -s -X POST "https://publisher.scrappey.com/api/v1?key=YOUR_API_KEY" \
    -H "Content-Type: application/json" \
    -d "{\"cmd\": \"request.get\", \"url\": \"$url\"}" &
done
wait
```

If you exceed your allotted threads, additional requests queue until a thread frees up. Keep your concurrent count at or below your limit to avoid delays.