---
id: livekit-plugins-sarvam
version: "1.6.10"
license: Apache-2.0
license_treatment: permissive
maintenance: active
---

# livekit-plugins-sarvam — Agent Framework plugin for services using Sarvam.ai's API.

License: permissive · Maintenance: active · Downloads: 91.8K/mo

## What it is and what it does

This package extends LiveKit Agents with Sarvam.ai's voice and language models, enabling speech-to-text conversion via Saarika models, text-to-speech synthesis via Bulbul models, and LLM chat completions with models like sarvam-30b and sarvam-105b. It is designed for developers building real-time conversational agents that need to support Indian languages and Indic scripts.

The plugin provides OpenAI-compatible interfaces for the LLM component, including tool-calling support, and integrates directly into the LiveKit Agents framework. It depends on livekit-agents and numpy, and requires only a Sarvam.ai API key to operate. The package is actively maintained and released frequently.

Use it for:

- Build multilingual voice agents that handle Indian-language speech input and output in real-time video/audio calls.
- Add Sarvam's LLM models to LiveKit Agents for chat-based conversational AI with tool calling.
- Convert user speech in Indian languages to text for processing by downstream agents or services.
- Synthesize responses back to users in Indian languages using Sarvam's TTS models.
- Deploy OpenAI-compatible LLM endpoints using Sarvam's 30B and 105B models within LiveKit workflows.

## Worth the install?

AI-flagged interpretation of the facts above — verify before relying.

Integrates Sarvam.ai's Indian-language voice AI services (speech-to-text, text-to-speech, and LLM chat) into LiveKit Agents for real-time conversational applications.

Yes. The package is actively maintained, has low install friction, carries a permissive license, and fills a clear niche for developers needing Sarvam.ai integration with LiveKit Agents. No known vulnerabilities. Install if you are building real-time agents that require Indian-language voice support or Sarvam's LLM models.

## Install

pip install livekit-plugins-sarvam
uv add livekit-plugins-sarvam
poetry add livekit-plugins-sarvam

## Installing livekit-plugins-sarvam

Before you install: Low friction install with a pure-Python wheel. Actively maintained as of 2026-08-14 with recent release history; requires Python 3.10 or later and depends only on livekit-agents and numpy.

License in practice: Apache-2.0 permissive license allows commercial and private use with minimal restrictions; suitable for most production deployments.

Quickstart:

pip install livekit-plugins-sarvam

from livekit.plugins import sarvam

llm = sarvam.LLM(model="sarvam-30b")

Requires a Sarvam.ai API key set as the SARVAM_API_KEY environment variable; Python 3.10 or later required.

Verify before relying:

- Whether STT and TTS integrations require additional setup beyond the API key beyond what the description states.
- Specific latency or throughput characteristics for real-time voice applications.
- Whether tool-calling support is fully compatible with all LiveKit Agents workflows.

## Package facts

- License: Apache-2.0 (permissive)
- Python support: supports_current
- Install friction: low
- Maintenance: active
- Downloads: 91.8K/month (top 15,000 on PyPI)
- Known vulnerabilities: none known

## Tags

sarvam.ai integration livekit, indian language speech to text, sarvam stt tts plugin, livekit agents sarvam, indic language voice ai, sarvam llm chat completions, real-time voice agents, voice-ai, indic-languages, realtime-agents

[View on SkillFed](https://skillfed.io/packages/livekit-plugins-sarvam) · [View on PyPI](https://pypi.org/project/livekit-plugins-sarvam/)