We attempted to create a server‑side WebRTC call using POST https://api.vapi.ai/call with { "type": "webCall", "assistantId": "f8bb22f6-7f63-4840-892a-ed3019bb12d7" } using our private API key (sk-). The API consistently returns: 400: type must be one of: outboundPhoneCall, inboundPhoneCall This suggests server‑side WebRTC call creation is not enabled for our org (or not supported). Could you please clarify: Is server‑side WebRTC call creation (type: "webCall") supported today, and if so, can it be enabled for our org? If not supported, what is the supported way to obtain call.monitor.controlUrl for WebRTC calls so we can perform live assistant updates (e.g. startSpeakingPlan changes) mid‑call? If live mutation of assistants during WebRTC calls is not currently possible, can you confirm that limitation explicitly? This capability is required for real‑time pacing control in our product.

Hi there, Thank you for your message. Our team is currently out of the office. We operate Monday through Friday, from 9:00 AM to 8:00 PM Pacific Standard Time (PST). We’ll get back to you as soon as possible during our normal business hours. If your message is urgent, please mark it accordingly or include “URGENT” in the subject line, and we’ll do our best to respond promptly. Warm regards, Vapi Customer Support Team

Im abit confused about the answers I got from the Composer, then the AI on the documentation page. Main question is: can II perform live assistant updates (e.g. startSpeakingPlan changes) mid‑call?

Hi, Direct mid-call assistant configuration updates are not currently supported. Once a call begins with a specific assistant configuration, properties such as startSpeakingPlan, model, voice, firstMessage, and system prompt remain fixed for that assistant instance for the duration of the call. This can sometimes cause confusion because assistant overrides are supported during call initialization, but changing the configuration after the call has already started is not supported. What you can do instead during a call: • Assistant handoffs using

assistantOverrides

via WebSocket control messages — this effectively switches to a new assistant instance with different settings. •

say

make the assistant speak specific text immediately. • `add-message`add messages to the conversation context. •

end-call

terminate the call. If you’d like help implementing assistant handoffs or WebSocket control messages, please let us know and we’ll be happy to assist. Best regards, Oshi Vapi Support