Compare realtime APIs

Pick up to four APIs from any category and see them side by side.

API 1API 2API 3API 4

	ElevenAgents (formerly Conversational AI / Agents Platform) ElevenLabs	xAI Grok Voice Agent API xAI	AssemblyAI Voice Agent API AssemblyAI
Category	Voice-to-voice	Voice-to-voice	Voice-to-voice
Status	GA	GA	GA
Est. per minute	$0.08 - 0.13	$0.08	$0.075
How that was worked out	$0.08 platform rate plus LLM pass-through; a third-party guide puts LLM at about $0.0005/min (GPT-5 Nano) to $0.045/min (GPT-5.5). Telephony and burst pricing extra.	Flat $0.08 per minute on grok-voice-think-fast-2.0 regardless of context length. Tool calls and text inputs are extra. xAI does not state whether minutes are session wall-clock time or audio time, so assume wall-clock time including silence.	Flat all-inclusive rate; telephony not included.
Pricing model	per-minute	per-minute	per-minute
Free tier	15 call minutes per month, 4 concurrent calls.	None documented for the voice agent.	$50 free credits at signup, no card.
Connects by	WebSocket, SIP, Twilio, Web/mobile SDKs, Embeddable widget	WebSocket	WebSocket
Audio in	pcm_8000/16000/22050/24000/44100/48000 or ulaw_8000 (negotiated per agent)	audio/pcm at 8000, 16000, 22050, 24000 (default), 32000, 44100 or 48000 Hz; audio/pcmu, audio/pcma (G.711); audio/opus. JSON base64 or binary transport.	PCM16 mono 24 kHz (audio/pcm default)
Audio out	Same enum as input	Same format options as input; speed 0.7 to 1.5.	PCM16 24 kHz (reply.audio)
Languages	Docs give both '31 languages' and '70+ languages' for the voice layer (inconsistent).	Docs say every voice can speak every supported language; TTS lists 20 languages and STT 38+. No explicit list for the voice agent.	Declared via input.language_codes or auto-detected.
Latency (vendor claim)	Described as low-latency; no figure on the overview page.	Vendor claim: sub-second latency.	Not stated on the pages read.
Key limits	Concurrency 4 (Free) to 40 (Business) concurrent calls; burst to 3x at double price Some LLMs are unavailable when EU data residency is enabled Call time measured on connection duration, not speech time	Session duration and concurrency limits are not published Resumption history dropped after 30 minutes of inactivity keyterms: max 100 terms, 50 characters each Ephemeral client secrets: expiry set by expires_after.seconds (example 300); session and anchor fields not supported	Session has a maximum duration (expires_at in session.ready); session_expired arrives with no advance warning Streaming concurrency on the free tier: 5 new streams per minute (general streaming figure)
High-severity warnings	LLM is not in the $0.08 Connection time, not talk time	Alias moved and price rose	None
Compliance	Enterprise: DPA/SLA, BAA for HIPAA, custom SSO; regional data residency hosts (EU, India, Singapore). SOC 2/GDPR not re-verified this session.	xAI states SOC 2 Type II, HIPAA eligible with a BAA, GDPR with EU data residency options, and that audio is never stored or used for training (vendor claims).	Vendor describes the API as PCI-certified; other certifications not re-verified.
Self-hostable	No	No	No
Last checked	2026-10-10	2026-10-10	2026-10-10
Key numbers and features
Flat $/min	$0.08	$0.08	$0.075
Audio in $/1M tok	-	-	-
Audio out $/1M tok	-	-	-
Free tier	Yes	No	Yes
Free credit $	-	-	$50
Native S2S	No	Yes	No
Tools	Yes	Yes	Yes
Image in	-	-	-
Own LLM	Yes	No	-
Voices	5,000	-	-
Cloning	Yes	Yes	-
Latency ms	-	-	-
Languages	31	-	-
Context tokens	-	-	-
Max session min	-	-	-
Concurrency	6	-	-
WebRTC	-	No	No
WebSocket	Yes	Yes	Yes
Phone / SIP	Yes	No	No
HIPAA	Yes	Yes	-
SOC 2	-	Yes	-
EU data	Yes	Yes	-
Open weights	No	No	No
High warnings	2	1	0