Sarvam Models APIs
Sarvam · Operations & Productivity
Certification per Microsoft Marketplace.
Evidence tier Source Confirmed · 9 captures on record
What the publisher says
As described on Microsoft Marketplace.
Unlock the Power of Indian Languages with Sarvam AI
What are Sarvam's model offerings?
Show the rest of the publisher’s description (25 more lines)
- Saaras (Speech-to-Text with Translation): Sarvam's speech-to-text model combines speech recognition and translation, allowing users to convert spoken language directly into translated text. It features automatic language detection, built-in context awareness for superior speech recognition accuracy, and direct English output, making it ideal for voice-based LLM applications.
- Mayura (Translation): Sarvam's colloquial translation model is trained on a highly diverse and high-quality dataset, enabling complex and contextual information to be made accessible across India. It supports both formal and colloquial translation across 10 Indian languages.
- Bulbul (Text-to-Speech): Sarvam's text-to-speech model converts written text into spoken words. It currently supports 6 unique Indian voices and has been trained for superior performance on domain-specific terms and special entities like symbols, currencies, and alphanumeric characters.
How are they different?
- Available in 10 Indian languages: All the models are available in 10 Indian languages & English.
- Best-in-class accuracy in Indian languages: The models are trained on highest quality dataset representing India's diverse and rich linguistic landscape making them superior in accuracy and performance.
- Contextual Understanding: These models are trained to understand and process colloquial language, idioms, and context-specific expressions, making them very compatible for conversational applications
- High performance, low latency: Compute-efficient and cost-efficient by design, these models offer blazing fast inference making them an ideal choice for real time applications.
- Reliability and application first: These models have been trained for real world applications with a specific focus on special entities, numbers, abbreviations making them the reliable choice for population scale use-cases.
Common Applications
These models serve as powerful building blocks for a wide range of applications, including:
- Multilingual Voice Assistants
- Real-time Language Translation Services
- Accessible Content Creation for Diverse Language Speakers
- Automated Customer Support Systems
- Language Learning Platforms
- Voice-Enabled Navigation Systems
- Multilingual News Aggregators and Summarizers
- Localized E-commerce Platforms
- Cross-lingual Information Retrieval Systems
- Inclusive Digital Government Services
- Multilingual Audiobook Production
- Voice-Controlled Smart Home Devices for Indian Languages
- AI-Powered Language Dubbing for Media Content
By leveraging Sarvam's AI models as foundational components, developers and businesses can create innovative solutions that bridge language gaps and cater to India's linguistically diverse population.
Preview
2 images
Agent build and provenance
See the full provenance
The layer-by-layer build, the evidence behind each claim, the risk basis and the cross-marketplace links are open to any account. Some rows are disclosed, some the source leaves Unknown; a free account shows you which.
Compliance
- FedRAMPConfirmedNot listed90%, registry-checkedNo FedRAMP Marketplace entry matched this vendor's domain, checked 2026-08-27registry recordas observed 2026-08-27
Confirmed means matched to a public authoritative registry. Claimed means the vendor or its listing states it, not yet cross-checked. A framework not shown was not found in any source we hold, which is not evidence against it. Not listed means a scoped registry check found no match for this vendor's domain: a No is a scoped registry check, not a compliance judgment. Confidence bands: 95% domain-verified, 90% registry-checked, 80% self-attested, 70% weak signal. Self-attested items marked “vendor's site” are gathered from the vendor's own website and are not verified by us.
Vendor
External enrichment
Plans and pricing as listed
2 listed- Sarvam- Speech to Text Translation Diarization: $0.49 per hour
- Sarvam- Speech to Text with Translation Bulk: $0.33 per hour
- Sarvam- Speech to Text Streaming: $0.33 per hour
- Sarvam- Formal Translation: $0.22 per 10k characters
- Sarvam- Speech to Text with Translation Streaming: $0.33 per hour
- Sarvam- Colloquial Translation: $0.22 per 10k characters
- and 72 more meters
- Sarvam- Speech to Text with Translation Bulk: $0.33 per hour
- Sarvam- Speech to Text Translation Diarization: $0.49 per hour
- Sarvam- Speech to Text: $0.33 per hour
- Sarvam- Text to Speech Streaming: $0.16 per 10k characters
- Sarvam- Speech to Text Bulk: $0.33 per hour
- Sarvam- Text to Speech: $0.16 per 10k characters
- and 72 more meters
Sources
Publisher resources
5 linksLinked repositories
Unknown means this listing does not publish a repository. It is not a statement that the code is closed, and a linked repository is not a claim that the publisher wrote it: the registry computes that relationship privately and does not publish it.
Evidence risk is the share of the build you cannot see before you deploy, not a security rating. Sign in to see the layer-by-layer basis for this band.