Every API you need
From speech to text. LLMs to translation. One platform, one API key, one dashboard.
Sarvam 105B
Flagship multilingual chat LLM for enterprise-grade Indian language tasks.
Sarvam 30B
Fast, cost-efficient multilingual chat LLM with strong reasoning.
Text to Speech
Bulbul v3
Natural-sounding speech synthesis in 10+ Indian languages.
Speech to Text
Saaras v3
Accurate transcription tuned for Indian accents and call-center audio.
Sarvam Vision
Document digitization and image analysis for Indian documents.
Translation
Mayura v1
High-quality machine translation across 10+ Indian languages.
Transliteration
Convert text between Latin and native Indic scripts.
Language ID
Detect the language of any text snippet across Indian languages.
First API call in 5 minutes
Install the SDK, paste your key, run the code. That's it.
pip install sarvamai from sarvamai import SarvamAI client = SarvamAI(api_key="YOUR_API_KEY") response = client.text_to_speech.generate( text="Namaste, yeh Sarvam AI hai.", target_language_code="hi-IN", model="bulbul:v2", speaker="meera", ) with open("output.wav", "wb") as f: f.write(response)
Works with your stack
Pre-built integrations for voice infra, agent frameworks, and automation tools.
Enterprise-ready. Data stays in India.
Compliance, control, and data sovereignty. Not bolted on. Built in from day one.
No training on your data
Your API inputs are never used for model training. Zero data retention after processing unless you explicitly request it.
- Data deleted after processing by default
- Opt-in retention with configurable TTL
- Separate data and model training pipelines
- Full DPDP compliance
Deploy on your terms
All processing happens within India. No cross-border transfers. For regulated workloads, we support VPC and on-premise deployment.
- India-only data processing
- VPC and on-premise options
- Consent-based voice cloning
- Content safety filters built in
Security and governance
Every API call is logged and traceable. Role-based access, audit trails, and data residency controls built into the platform.
Best-in-class accuracy for Indian languages
Bulbul V3 outperforms global providers on character error rates.
Listener preference rate (8kHz)
Higher is better
ElevenLabs Flash V2.5
ElevenLabs V3 Alpha
Cartesia Sonic-3
Transparent pricing
Pay per use. Start with 1,000 free credits. No credit card required.
Start building with free credits
Start building with free credits