Hi Vapi support team, We have run into an urgent issue with vapi when using outbound calling assistants. Please see the 429 “Too many requests” error below. The current concurrency limit for the organization that experienced this error is 20 calls and we have been attempting 8 calls per minute so we have not hit any limits there. Can you please let us know what the issue is?

Hi there! 👋 I'm sorry to hear you're experiencing 429 "Too Many Requests" errors with your outbound calling assistants, especially since you're well below your concurrency limit of 20 calls. While I can't pinpoint the exact cause without specific details about your account, I can definitely help explain some potential reasons this might be happening. There are several possible causes for this error even when you're below your concurrency limit: 1. **System Errors in Concurrency Tracking**: According to [our documentation](https://discord.com/channels/1211482211119796234/1211483291191083018/threads/1326262542858850445), there might be a system error where calls aren't being properly cleared from your concurrency pool, causing the system to incorrectly think you've reached your limit. 2. **Stuck Calls**: Calls that are stuck in an "in-progress" state can occupy your concurrency limit, preventing new calls from being made. [Our documentation](https://discord.com/channels/1211482211119796234/1211483291191083018/threads/1334236814705299519) indicates that API downtime can cause calls to remain "In Progress," occupying your concurrency limit. 3. **Internal Vapi Issues**: There could be issues with our workers that cause calls to fail and result in concurrency limit errors. As noted in [our support threads](https://discord.com/channels/1211482211119796234/1211483291191083018/threads/1334057850787463231), "There was an issue within our workers because of this calls were failing and eventually resulted into concurrency limit 404." 4. **Rate Limiting vs. Concurrency**: While you're monitoring concurrency (simultaneous calls), the rate at which you initiate calls (8 per minute) might still trigger rate limiting. [Our documentation](https://discord.com/channels/1211482211119796234/1211483291191083018/threads/1328845842018340968) suggests implementing an artificial delay between calls. Here are some troubleshooting steps you can try:

Hey! To help track down this issue, could you share: \- The call ID \- When exactly this happened This would really help us figure out what went wrong!

I had thousands of failed calls without and no error logs on vapi dashboard. I'm not sure whoich of these are the callID, I can't see the failed call logs on the dash Session ID cm9d2e14z0q91f4p1azd0zy11 Chat ID cm9d2e14s0q8zf4p11hrbrs8u This has happened starting at 2025-04-10T19:14:04.325Z

Looking into it

Hey, could you share your organization ID?

here you go 41df130e-c321-46a1-b36f-b9f66247dfec

Hey Jasmine, can you try now? We've added rate limiting to prevent network state sync issues with the frontend and spam requests.

still not working, even when I try 1 call it never gets placed inbound works but outbounds are not going through no error logs in twillio either, so it's not on their end

Hey Jasmine, I just looked into your Org ID again, and under your Org ID, the API request plan has 500 requests per minute. Even if you touch this limit, it will get reset after two minutes. As of now, I don't have access to the memory where these are blocking keys are stored for rate limiting, so I cannot clear them manually. But in case as you are saying that you can't even make a single call, could you share a Loom video so that I can take a look and even share with my team? If considering 500 requests per minute and two minutes of expiration time, and if you are still unable to make the calls, then there might be something else. I know it's frustrating, but your patience and attention will help us resolve the issue for you.

I am experiencing rate limit errors as well. Org ID is dfa21930-04c6-4006-b84c-d7e1c96ae23f and even after +5m cooldown I'm getting the error (hope I'm addressing my issue correctly – don't want to hijack or misguide the issue that Jasmine is reporting. But I joined Discord for a faster response time and checked the search for similar issues. Sorry in case I misused this thread)

@Shubham Bajaj @User please see below 1. ALL requests to api.vapi.ai for the last 4 hours from production 2. each request POST request info and 3. error we’re getting from not being able to parse the “Rapid pace” response https://cdn.discordapp.com/attachments/1360314178128318635/1361392300713840670/screenshot_2025-04-14_at_20.png?ex=67fe96e1&is=67fd4561&hm=80b7b82acbb97caf81e9710c8d48f6430664c6ea5ffa7bfc482fafe436316d06& https://cdn.discordapp.com/attachments/1360314178128318635/1361392300965499012/screenshot_2025-04-14_at_20.png?ex=67fe96e1&is=67fd4561&hm=f0a805613597b9ba57f610e13a23400223ad520de001b2b16e05644ab0209a9f& https://cdn.discordapp.com/attachments/1360314178128318635/1361392301221613648/screenshot_2025-04-14_at_20.png?ex=67fe96e2&is=67fd4562&hm=b02e776d4459e021201e4c6f6c78afe07277920d741ceb525bd62a6250495e44&

Fixed.