Compare realtime APIs

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

Mistral Voxtral Mini Transcribe Realtime
Mistral AI
Sarvam AI Saaras Realtime STT
Sarvam AI
Voxist ASR (via OVHcloud / Scaleway marketplaces)
Voxist
CategorySpeech-to-textSpeech-to-textSpeech-to-text
StatusGAGAGA
Est. per minute$0.006$0.0059 - 0.0088$0.0063 - 0.016
How that was worked outFlat announced rate; not confirmed on a live pricing table in this pass.INR 30-45/hr converted at about INR 85 per USD (third-party FX note, July 2026); currency risk applies.Licence EUR 0.35-0.88/hr (~$0.006-0.016/min at ~1.1 USD/EUR) before GPU compute. Low confidence.
Pricing modelper-minuteper-hourper-hour
Free tierNot stated.INR 100 credits for new users.None stated.
Connects byWebSocketWebSocketWebSocket, REST
Audio inAudioFormat(encoding='pcm_s16le', sample_rate=16000) in docs examples; mic example sends 480 ms chunks.encoding and sample_rate parameters on the realtime channel (values not captured).WAV, mono only; examples use 16 kHz.
Audio outn/an/an/a
Languages13 languages (vendor).Indian languages plus English (exact list not captured).Not listed on the billing page (examples show en and French output).
Latency (vendor claim)Vendor claim: configurable latency down to sub-200 ms; press reports ~1-2% error rate at 480 ms delay (secondary source).Not captured.Not captured.
Key limits
  • rt_ client tokens expire after about 900 seconds; mint them just before connecting.
  • Python realtime needs SDK v2 (mistralai>=2, extra [realtime]); not available in v1 SDK.
  • See Credits and Rate Limits page (not captured).
  • WAV mono only.
  • Streaming has no diarization.
High-severity warnings
  • None
  • None
  • None
ComplianceNot re-verified. Self-hosting keeps audio fully in your infrastructure.Not re-verified.Runs in your own cloud project; certifications not checked.
Self-hostableYesNoYes
Last checked2026-10-102026-10-102026-10-10
Key numbers and features
$/hour$0.36$0.35-
Bills silence---
Free tier-Yes-
Free credit $---
Live speakersNo-No
Turn detect---
Keyterms---
PII redact---
PartialsYes--
Mixed langs---
8 kHz phone--No
Latency ms200--
Languages13--
Max session min---
Concurrency---
WebRTCNoNoNo
WebSocketYesYesYes
gRPCNoNoNo
HIPAA---
SOC 2---
EU data--Yes
Self-hostYesNoYes
Open weightsYesNoNo
High warnings000