Sarvam AI: Saaras v3

Overview

Sarvam AI's speech recognition model, available beside v4 with the same five output modes (transcribe, translate to English, verbatim, transliterate, code-mix) for 22 Indian languages and Indian English, with real-time streaming, and offered for self-hosting on Amazon SageMaker.

Provider

Sarvam AI

Modality

Transcription

Input

audio

Output

text

Weights

Closed

Distillable

NoLicence or terms, checked October 9, 2026.

Tags

speech-to-text, speech-translation, audio, indian-languages

Docs

Provider documentation

Data policy

Prompts go to

Sarvam AI, on its managed environment in Microsoft Azure's Central India region. Self-hosted on Amazon SageMaker in your own AWS account, audio and documents stay in your account.

Retention

Sarvam keeps Model API data until the workspace Owner sets a retention period, and No retention (0 days) is available for Model APIs. Sarvam's terms let it train its own models on inputs and outputs as its Privacy Policy and the law allow, with consent where required, and its Trust Center says one customer's data never trains models for another customer.

Zero data retention

Available

Region

India: Sarvam's managed production environment runs in Microsoft Azure's Central India region.

Policy

Sarvam AI data policy

Summarised from the provider's published terms. The provider's own policy is what applies.

Where you can run it

How Saaras v3 can be deployed, bought and reached through Safeguard.

Deployment

Public cloudPrivate cloud

Procurement

Not published yet

Inference regions

🇮🇳IndiaCloud provider's availability page, checked October 9, 2026.

Closed weights stay with the provider, so this model runs in the provider's cloud, never on-prem or air-gapped.

Pricing

Notes

Billed per second at INR 30 per hour of audio, INR 45 per hour with speaker diarization (batch). Prices are in Indian rupees only, so no US dollar price is shown.

Source

Sarvam AI pricing , last verified October 9, 2026

No markup on model usage: you pay the provider's list price.

API

OpenAI-compatible. Your API key and access to this model are set up for your workspace when you choose Use with Safeguard.

from openai import OpenAI

client = OpenAI(
    base_url="https://api.safeguard.sh/gpt/v1",
    api_key="SAFEGUARD_API_KEY",
)

response = client.chat.completions.create(
    model="sarvam/saaras-v3",
    messages=[{"role": "user", "content": "Hello"}],
)
print(response.choices[0].message.content)

Base URL https://api.safeguard.sh/gpt/v1, model sarvam/saaras-v3.