[Docs] Added advanced concurrency limits section

fern/advanced/concurrency-limits.mdx

@@ -0,0 +1,38 @@
+---
+title: Managing Concurrency Limits
+subtitle: Tips for managing concurrency limits
+slug: advanced/concurrency-limits
+---
+
+# API Retry and Concurrency Management
+
+## Handling Retries
+Retries are effective for API calls of short durations. For longer-running calls, such as those exceeding 10 minutes, ensure your retry intervals align with the call duration to prevent premature attempts or wasted resources.
+
+## Concurrency Limits
+The API enforces a concurrency limit, which governs the number of simultaneous API calls. When this limit is exceeded, the API responds with:
+```http
+HTTP/1.1 400 Bad Request
+{
+  "error": "Over Concurrency Limit"
+}
+```
+## Avoiding Concurrency Limit Breaches
+
+High volumes of outbound calls, especially when queued via the dashboard, can lead to hitting the concurrency cap. Follow these best practices to avoid such breaches:
+
+- **Reduce simultaneous calls:** Minimize concurrent requests to keep within the limit.
+- **Introduce delays:** Space out call initiations to manage traffic.
+- **Optimize call durations:** Ensure calls complete efficiently to prevent queue overloads.
+
+## Managing Concurrency Effectively
+
+For enterprise plans, additional concurrency capacity can be purchased. Use the monitoring tools provided in the API dashboard to:
+
+- **Track active call usage:** Monitor ongoing API requests to stay within limits.
+- **Analyze and adjust thresholds:** Identify patterns and reconfigure settings as needed.
+- **Reference your subscription schema for limit details:**
+  - `Subscription.concurrencyLimitIncluded`: Default concurrency capacity.
+  - `Subscription.concurrencyLimitPurchased`: Additional purchased capacity.
+
+For more details, refer to the [Subscription schema](https://api.vapi.ai/api/).

fern/docs.yml

@@ -251,6 +251,8 @@ navigation:
             path: voice-fallback-plan.mdx
           - page: OpenAI Realtime 
             path: openai-realtime.mdx
+          - page: Managing Concurrency Limits
+            path: advanced/concurrency-limits.mdx
       - section: Glossary
         contents:
           - page: Definitions