Bind voice agents
- Bind an agent for each direction where you want normal AI call handling.
- You can assign different inbound and outbound agents to the number.
- Leave an agent unset to disable normal AI handling for that direction. For example, if you only make outbound calls and don’t want callbacks, leave
inbound_agent_idunset.

Bind inbound and outbound agents to your number
Handle inbound overflow
Inbound calls count toward your workspace concurrency limit. To keep inbound calls available during high outbound volume, you can reserve part of your concurrency for inbound calls. See Reserved Inbound Concurrency. If all concurrency slots are in use, Retell briefly waits for a slot to open. If no slot opens after about 40 seconds, Retell transfers the call to the phone number’sfallback_number when one is
configured. Without a fallback number, the call ends with concurrency_limit_reached.
Divert calls with Backup mode
If you need to pause AI handling on one number, use Backup mode to send new inbound calls directly to that number’s fallback destination. For example, route calls to your staffed support line while you investigate an agent issue.- Open Phone Numbers and select the number.
- Under Fallback Number, enter and save the destination number. Backup mode stays disabled until a fallback number is configured.
- Turn on Backup mode and confirm the change.
Fallback billing
Both Backup mode and automatic concurrency fallback route the caller to the fallback destination without an agent handling the call, so no agent fee applies. For Retell-managed numbers, telephony charges apply at the rate for the fallback number’s country, from when the destination answers until the call ends (as of November 2026). Fallback cost = connected seconds ÷ 60 × the fallback destination’s per-minute rate. At an example rate of $0.015/min, a 10-minute fallback conversation costs $0.15, with no agent fee. See rate assumptions and examples. Ano_concurrency_fallback call record can show that the agent call has ended while the fallback conversation is still accruing telephony charges. Backup mode can also incur telephony charges without creating an agent call record. With custom telephony, your provider bills the phone connection directly; Retell does not add a telephony fee. See call duration and telephony charges.
A/B testing
See A/B Testing.Inbound call webhook
You may want to use different agents for inbound calls to the same number, or provide dynamic variables and other per-call fields. Read more at Inbound Call Webhook.Route calls to Retell’s SIP endpoint
If you’re routing inbound calls from your own carrier or PBX, point your SIP trunk or dial plan at Retell’s SIP server:- SIP server URI:
sip:sip.retellai.com - Transports: TCP (recommended), UDP, TLS, mTLS. Append
;transport=tcp,;transport=udp, or;transport=tlsto the URI to select one. - Media encryption: SRTP (requires TLS transport)
- Audio codecs: PCMU, PCMA, G.722 (HD)
- Re-INVITE support: Retell answers in-dialog re-INVITEs (media re-anchor, codec change, SRTP re-key, hold/unhold, offerless) — see the custom telephony FAQ
- IP blocks to whitelist:
18.98.16.120/30(signaling only),3.42.144.0/23,153.57.128.0/18,143.223.88.0/21(certain US traffic),161.115.160.0/19(certain US traffic)
Inbound custom SIP headers
- You can use custom SIP headers to pre-set dynamic variables for a call.
- Retell extracts any header starting with
sip.h.x-, along with common SIP headers likeDiversion,History-Info,User-To-UserandP-Asserted-Identity, and converts each into a dynamic variable by stripping thesip.h.prefix.- E.g.
sip.h.x-caller: abc->x-caller: abc - E.g.
sip.h.p-asserted-identity: +12345678910->p-asserted-identity: +12345678910 - E.g.
sip.h.diversion: <sip:2000@192.168.254.254>;privacy=off;reason=no-answer;counter=1;screen=no->diversion: <sip:2000@192.168.254.254>;privacy=off;reason=no-answer;counter=1;screen=no
- E.g.
Get call detail
- API: Use the Get Call API to get information like transcript, recording, and latency tracking.
- Webhook: Set up webhooks to receive real-time updates when a call is initiated, ends, and is analyzed. Read more at Call Webhook Guide.

