Hi team, Thanks in advance, this one's a bit long --- Running into a critical issue where the agent is prompted to use the query tool to look up details for a procedure. When using the query tool I can see it's querying the right knowledge base However, there are either no results returned or the wrong knowledge base is used. I have another agent with the exact same setup and it consistently gets it right, while the one in production consistently fails. We don't have transparency into HOW the agent is querying, only that it queries a specific KB: Query:

{"knowledgeBaseNames":["clinic_appointment_type"]}

Incorrect result: `[ { "name": "qr_j573wcg4bba9gb6gfeef2dt8s97tcprf", "toolCallId": "call_yXsBEQ0qYn9QBbP8MJoFiybA", "type": "query", "result": "```text\nMonday: 08:00 - 18:00\n```" } ]` Correct Query: Query :

{"knowledgeBaseNames": ["clinic_appointment_type"]}

Correct result: `[ { "name": "qr_j976q1pdjjwgwgznnf96wmdrsn7tntr5", "toolCallId": "call_N9auENLGPuszgbNFjV0EmM8F", "type": "query", "result": "```json\n[\n {\n \"_creationTime\": 1762168946683.3276,\n \"_id\": \"jh79dwk5k1be9sh0mttbr27ty57tqag2\",\n \"agentId\": \"j976q1pdjjwgwgznnf96wmdrsn7tntr5\",\n \"estimatedDuration\": 30,\n \"name\": \"Check Up/Scale Clean\",\n \"pmsId\": \"Sp4DmVITUEyTAPKMPnqupw\",\n \"eventClassId\": \"Sp4DmVITUEyTAPKMPnqupw\",\n \"practitioners\": [\n \"Stephen Vincent Strange\",\n \"Bruce Banner\",\n \"Tony Stark\",\n \"Natalia Romanov\"\n ]\n }\n]\n```" }, { "name": "qr_j976q1pdjjwgwgznnf96wmdrsn7tntr5", "toolCallId": "call_lqUo0ydWYu5nY4z2uRymRqn4", "type": "query", "result": "```text\nClinic Opening Hours:\nMonday: 09:00 - 17:00\nTuesday: 09:00 - 17:00\nWednesday: 09:00 - 17:00\nThursday: 09:00 - 17:00\nFriday: 09:00 - 17:00\nSaturday: 09:00 - 17:00\nSunday: 09:00 - 17:00\n```" } ]` Broken Call ID: 019b02db-f100-7bb8-987f-57eea4fe632c Working Call ID: 019b02e2-a3d1-7006-8667-088991a183be

I have an assistant who confirms the order. I initially passed the order context in variables, but now I do it using a tool. The data I want to pass is JSON. { "jobId": "2ee8880a-4051-450b-ba02-8880697bc4df", "shippingAddress": { "firstName": "Jan", "lastName": "xxxx", "address1": "Warszawska 100A/1", "city": "Katowice", "postalCode": "00-001", "countryCode": "PL", "phone": "+xxxxxxxxxxx" }, "items": [ { "sku": "M1", "name": "The Multi-location Snowboard", "qty": 1, "price": 693.45, "variantId": "47466035151073", "productId": "9198231847137" } ], "totals": { "subtotal": 642.08, "tax": 51.37, "grandTotal": 693.45 }, "productInstructions": { "replacements": [ { "toSku": "P2", "reason": "Brak na magazynie", "fromSku": "M1" } ] }, "store": { "id": "store_98rur3437", "platform": "shopify", "name": "Fashion Store", "domain": "fashion.example" }, "order": { "orderId": "6749360455905", "orderNumber": "1025", "createdAt": "2025-11-23T08:27:10+00:00", "updatedAt": "2025-11-23T08:27:10+00:00", "currency": "USD", "financialStatus": "pending" }, "customer": { "customerId": "8751628648673", "email": "karine.ruby@example.com", "firstName": "Jan", "lastName": "Kowalski", "phone": "+xxxxxxxxxxx", "fullName": "Jan xxxxxxxxxxx" }, "customerName": "Jan Kowalski" } But in both cases. After receiving it, the system triggers the first message. Then there's a 10-second pause, sometimes longer... and then it says its own thing. It's always the same. Calling the endpoint and returning the JSON takes a while. Can someone help me solve this?

Hi, I’m experiencing an issue with your TTS system when using Turkish as the selected language. Even though the voice and language settings are correctly set to Turkish, the model incorrectly pronounces numbers and mathematical expressions in English. For example: “3+1” is spoken as “three plus one” instead of “üç artı bir” Similar issues happen with other numeric expressions, dates, measurements, and symbols. After investigating, it appears the problem is related to the text normalization layer: Numbers and operators seem to be normalized using an English-centric or universal normalization pipeline, which overrides the intended Turkish locale. This results in incorrect verbalization before the TTS voice model processes the text. Could you please clarify: Does your TTS engine support locale-specific numeric normalization for Turkish (tr-TR)? Is there a way to force the locale for number and symbol verbalization? Can text normalization be disabled or set to verbatim mode? Is there an upcoming update that includes Turkish-specific text normalization rules? I would be happy to provide test samples, logs, or audio outputs if needed. This issue is important for my AI call project, as incorrect number pronunciation affects the naturalness and accuracy of the call flow. Thank you in advance, looking forward to your support.

Sip inbound calls are ending in 30 seconds. Today we are facing this problem. Are there anybody has the same issue?

Hi Vapi Support, I am building an AI agent specifically designed to navigate IVR phone trees. I am running into a critical failure case with IVRs that loop continuously without pausing (specifically the Costco customer support line). Call ID for investigation: 019b025e-784a-7ff6-b98a-a40e418f9fc5 Timestamp: 09 Dec, 2025, 14:08 The Issue: The Costco IVR reads a very long list of hours and then loops the menu options repeatedly without any silence gap between the end of the menu and the start of the loop. Because the IVR never stops speaking for more than a fraction of a second, the agent's VAD (Voice Activity Detection) or Endpointing logic never triggers the "User Finished Speaking" state. Consequently, the LLM is never called to process the transcript, and the agent sits there recording the loop indefinitely until the call fails or times out. What I have tried: I have updated the System Prompt with strict instructions to "Interrupt immediately upon detecting a repeated phrase" and "Do not wait for silence," but this hasn't worked. I suspect this is because the System Prompt is only evaluated after the turn is considered complete (which never happens due to the continuous audio). My Questions: Is there a configuration setting to make the agent process the transcript stream in real-time and trigger a tool while the other party is still speaking? Can I adjust the VAD silence threshold to be more sensitive (e.g., trigger on < 500ms gaps)? Is there a way to force an interruption or "barge-in" programmatically if the user/IVR has been speaking for X seconds without a break? Example Transcript (Looping): "...To reach the pharmacy press 4. To reach the administrative staff press 1. For warehouse hours... [Loop continues instantly]" Any guidance on how to handle continuous audio streams without silence gaps would be appreciated. Thanks.

Hi, I’m reaching out regarding an issue we’re currently experiencing with VAPI tools. When we create new tools, some of the existing ones suddenly disappear from the interface, as if there were a limit to the total number of tools allowed on the platform. After checking the documentation, we didn’t find any mention of such a limitation. Have you already encountered this type of behavior on your side, or is there any known constraint we should be aware of? Unfortunately, it’s difficult for us to show the phenomenon directly, as it happens intermittently. Thank you in advance for your help.

When will this be implemented? Can you suggest alternatives? Or Is your Knowledge Base working efficiently now? https://cdn.discordapp.com/attachments/1447855446294659102/1447855446529675408/Screenshot_2025-12-09_153624.png?ex=693923f6&is=6937d276&hm=e6821d4e49a42ab56bd48301054e9c91b4de9b6d187764abfa9b11bb329bf076&

All of tools for my inbound agent stopped working. I have been trouble shooting for hours, I can’t figure it out. My whole agent worked yesterday now none of my tools work. I have plenty of credits I have no idea. Has anyone had this happen? #1211483291191083018

I have successfully configured the AI Voice agent to make outbound calls and receive inbound calls to/from external numbers through the SIP Trunk. I would also like the AI Agent to be able to call/transfer calls to internal numbers on the SIP Trunk PBX. When transferring calls using the transfer call tool. If I use the SIP Trunk number and extension, it goes to the main entry point to the PBX and the extension is ignored. If I use just the extension number as a phone number, it uses that as if it is an external number and the call fails. Is there some additional configuration I should be aware of to route AI Agent transfer calls to internal numbers on my PBX (VoIPline)? I've contacted my PBX provider and they indicated they didn't see a request to transfer the call, so it seems to be coming in as a new call? From their support: "According to the logs, we can see the call was answered by the SIP Trunk, but we didn't see any evidence of a call transfer"

I am trying to initiate an outbound call but the API is giving me an error with the number format. I checked the number and its fine. It seems like there is an issue in the way VAPI validated the number. This is a number from Brasil The number format is: +55 xx xxxx xxxx

Anyone experiencing the same? #conference

Grok 4.1 Fast is available from some weeks ago and would be nice to have access to it alongside all the upcoming Grok new LLMs, which are SOTA at function calling. If you also want this, please upvote the feature request 👇 https://vapi.canny.io/feature-requests/p/grok-41-fast-non-reasoning-reasoning

My payment method is not going through but the information is correct, how can I if this?

| ID | Start Time (UTC) | End Time (UTC) | |--------------------------------------|-------------------------------|-------------------------------| | 019af829-bdbb-744b-82de-b11e874a74a2 | 2025-12-07 09:34:36 GMT+0 | 2025-12-07 09:35:32 GMT+0 | | 019af82a-b35a-7113-b889-9e37c7913073 | 2025-12-07 09:35:39 GMT+0 | 2025-12-07 09:36:34 GMT+0 | | 019af82a-bdb0-7881-b722-6b280ded721c | 2025-12-07 09:35:42 GMT+0 | 2025-12-07 09:36:36 GMT+0 | | 019af82a-c7c7-7113-a045-1cab63f6edbb | 2025-12-07 09:35:44 GMT+0 | 2025-12-07 09:38:13 GMT+0 | | 019af82a-d270-7113-b889-a70d3b621196 | 2025-12-07 09:35:47 GMT+0 | 2025-12-07 09:36:15 GMT+0 | | 019af82a-e8bc-7881-b722-84eca0fdc8fd | 2025-12-07 09:35:53 GMT+0 | 2025-12-07 09:37:43 GMT+0 | | 019af82a-fbc3-7119-88e7-90516a16db77 | 2025-12-07 09:35:57 GMT+0 | 2025-12-07 09:36:59 GMT+0 | | 019af82b-168a-7003-8ead-b94d271eded6 | 2025-12-07 09:36:04 GMT+0 | 2025-12-07 09:36:56 GMT+0 | | 019af82b-aa98-7003-8ead-eee73349633a | 2025-12-07 09:36:42 GMT+0 | 2025-12-07 09:37:14 GMT+0 | | 019af82b-c452-7eef-956f-29a0c1dd22ae | 2025-12-07 09:36:49 GMT+0 | 2025-12-07 09:37:49 GMT+0 | at 2025-12-07 9:38 TIME STAMP I have trigger 4 calls and all of them failed with concurrecy issue, but at that time the available were 7. The yellow ones have been completed by the said timestamp, where the error has been thrown. Folks Please Help me understand If I missed something or fix this issue. Thank you much!! 👀 org ID : 3ba2859e-36e2-4306-be9a-bbfd42ed38a6 https://cdn.discordapp.com/attachments/1447670493963358208/1447670494437576725/Screenshot_2025-12-09_at_12.53.46_AM.png?ex=693877b6&is=69372636&hm=f8632a4444f9ce264a68139a5deb0794a28808df8503814c8e5dbd1c15151e98&

Hi everyone, I’ve been trying to integrate RingCentral with Vapi so that calls arriving on a specific RingCentral extension can be forwarded to a Vapi AI agent, but I’m running into difficulties getting it to work. I tried using SIP trunk configuration in Vapi following the SIP docs, but it doesn’t seem to work with RingCentral. From an older thread I found, Vapi doesn’t support direct SIP registration with RingCentral and recommends using an intermediary SIP provider like Twilio or Plivo. I’d really appreciate clear answers to the following: Is it possible to integrate RingCentral directly with Vapi via SIP? Or is an intermediary SIP provider required to make this work? Is there any specific documentation / step-by-step guide for this setup? Thanks in advance!

Hey guys, So i have created a squad for my project which consists of multiple assistants for supporting different languages. For switching between models i created handoff tools which should only switch if user explicitely asks to switch the language model. But for some reason it switches mid conversation after it ask user's age. I can't seem to find the issue. Can someone help me out with this one? @User @Vapi Support Bot Browser: Chrome, Brave https://cdn.discordapp.com/attachments/1447658005934899372/1447658009416433847/image.png?ex=69386c15&is=69371a95&hm=09e63485e1870ca9710328dd4c0ca992d9314c08518c52317d14c1b0c88e5829&

I was trying to add a ai chat widget to my site and others. According to what I seen there is an actual tab that says widget that lets you customize the appearance etc. This tab is not showing up on my account dashboard. Can you please enable this tab or let me know how I enable it.

When an org is deleted any number that was assigned seems to remain in the system, preventing the number from being imported again to an new org. So if you accidentally delete an org with a number and want to create a new org and import the old number you cannot. This is causing us significant issues at the moment. We import numbers from Vonage and even in Vonage I can see the connection to VAPI being successfully created but VAPI does not allow the phone number to be added to the org. Surely the number relationship should be deleted when the org is deleted. Please help!

Greetings, I would like to store the audio recording on my s3 bucket. I know the vapi dashboard has integration with s3 but the dashboard configuration causes several files to be upload per 1 single audio call (stereo,mono etc...) and I want to save only the original audio. Is there any settings that can control what vapi uploads to my s3 so that only the original audio is saved as I'd like to avoid having extra data uploaded and deleting it myself.

Trying to run the react sdk results in

Uncaught TypeError: _vapi_ai_web__WEBPACK_IMPORTED_MODULE_2__ is not a constructor

Issue was created here https://github.com/VapiAI/client-sdk-react/issues/35 There's a simple reproduction repo

Is anyone else seeing "The transcript was not provided" for voice tests about 50% of the time? https://cdn.discordapp.com/attachments/1447607084182143067/1447607084417028116/Screenshot_2025-12-08_at_8.13.20_AM.png?ex=69383ca8&is=6936eb28&hm=6f187ae88317ffd5dfddf8ac93ba95050c176983a56f75ad88a2fbadf5c2b7d2&

Hi there! One of my calls failed with the issue all.in-progress.error-providerfault-openai-llm-failed. This is the link to one of the examples https://dashboard.vapi.ai/calls/019af9e7-7c8c-7aa6-9b95-891cf1144814 Is this a bug? https://cdn.discordapp.com/attachments/1447606901721272373/1447606902203613195/Screenshot_2025-12-08_at_8.11.14_AM.png?ex=69383c7c&is=6936eafc&hm=487583c9b8b5c335137c738251143b9cd141faf77ed232013eab726b6c5a4fb2&

Dear Vapi Team, I am a customer of the VoIP provider DIDlogic and am currently trying to use the “Bring Your SIP Trunk Carrier” integration. During the setup, I encountered an issue: when I enter the FQDN of the DIDlogic SIP gateways (sip.lax.didlogic.net or sip.nyc.didlogic.net), I receive the following error: “Failed to create SIP trunk credential.” However, if I replace the FQDN with the corresponding IPv4 address, the SIP trunk is created successfully without any errors. Back in September 2025, I set up Vapi for another project and did not experience this issue. Could you please investigate and address this problem? Thank you in advance for your assistance. https://cdn.discordapp.com/attachments/1447585163000156192/1447585163780554913/0.png?ex=6938283d&is=6936d6bd&hm=e8275a96278b634e6a40b2e3658e5838a9a017f573a65e5ca02a8de433f772a3&

We are using the realtime model of chatGPT, the customer complained to us that the AI is not speaking correctly, can you identify the issue and give us some step to fix it.

Please help me to know which other voice services will support ssml tags other than 11 labs.

Hi! So, I'm building a software for scheduling calls for agencies, and I've worked pout my workflow for my entire coding operation, up till the point of making proper phone calls, among other functionalities being configured properly. Upon testing the created outbound calls, the latency seems a bit too much for the outbound calls, its very normal for inbound calls, and web calls however. How do I fix that?

Please help me to know which other voice services will support ssml tags other than 11 labs.

Wanted to clarify 3 things from the docs **1. Use HIPAA-compliant accounts with all third-party providers (STT, LLM, TTS). **Does this just means we must select the right providers (i.e. Deepgram, OpenAI, Vapi) for our assistant, right? Do you handle BAAs with those orgs? I don't want to go wrangle BAAs & update accounts with all of them. 2. Tool Calls / Integrations. How does integrating with other services work with tool calling? We might be interfacing with external providers to accomplish our work over the phone (i.e. we get a patient reference number and pull in specific data for that patient) and need to be able to execute tool calls while on the phone. 3. End of Call workflows. Our server is compliant. We need to get end of call reports w/ transcripts to understand what happened, process the data, and share it with our clients. Is that functionality still supported in a HIPAA-enabled instance?

Dear Vapi Support Team, I would like some assistance on configuring my inbound calls using my own SIP trunk (BYOC) with Vapi.ai to leverage my existing PBX (Yaxxa in this instance in NZ). I have tried following the steps provided in the documentation and have also connected with Yaxxa to no avail. Would love to connect with one of the engineers to know what would be the issue when connecting my own SIP Trunk.