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

# Rate limits

> Default rate limits for deployments and inference, and how to request higher throughput

SynapsAI Cloud applies rate limits to protect platform stability. Limits may vary by account tier — contact [support](/support/contacting-support) to discuss higher limits.

## Model deployment rate limits

By default, each account can deploy **one model at a time**. Contact support if you need additional concurrent deployments.

## Model inference rate limits

Each deployed model supports up to **100 requests per second (RPS)** by default. Contact support for higher RPS requirements.

## Model instance rate limits

A deployed model can run on multiple instances. Each instance typically handles **6–8 concurrent requests**, depending on model type and size. The autoscaler adds instances when per-instance load exceeds your [scale up threshold](/manage/autoscaling#scale-up-threshold).

## HTTP 429 responses

When rate limits are exceeded, the API returns HTTP `429 Too Many Requests`. Implement exponential backoff with jitter in production clients:

```python theme={null}
import time
import random

for attempt in range(5):
    try:
        response = client.chat.completions.create(...)
        break
    except Exception as e:
        if "429" in str(e) and attempt < 4:
            time.sleep((2 ** attempt) + random.random())
        else:
            raise
```

## Related

* [Autoscaling](/manage/autoscaling) — scale instances to absorb higher concurrency
* [API overview](/api-reference/introduction) — error codes and authentication
* [Troubleshooting](/support/troubleshooting) — diagnose failed requests