Getting Started

MODELS

All available models on CallMissed β€” Sarvam AI for STT/TTS/LLM, and 300+ text models via OpenRouter. All accessible through one OpenAI-compatible API.

Overview

CallMissed provides access to two model providers through a single OpenAI-compatible API:

  • Sarvam AI β€” purpose-built for Indian languages. STT, TTS, and LLM optimized for Indic languages (Hindi, Tamil, Telugu, Bengali, and 18 more).
  • OpenRouter β€” 300+ text models from every major provider (OpenAI, Anthropic, Google, Meta, Mistral, DeepSeek, and more) through one endpoint.

All models use the same authentication and request format. Just change the model field.

Sarvam AI Models

Speech to Text

ModelDescriptionLanguages
saaras:v3Latest STT β€” best accuracy22 Indic + English

Text to Speech

ModelDescriptionVoices
bulbul:v3Natural TTSmeera, kalpana, arvind, amol

Chat Completion (LLM)

ModelContextBest For
sarvam-30b32K tokensGeneral purpose, Indic languages
sarvam-105b32K tokensFlagship β€” highest quality

OpenRouter Models

Access 300+ models via the same /v1/chat/completions endpoint. Use the OpenRouter model ID as the model field.

Model IDProviderInputOutputContext
openai/gpt-5.4OpenAI$2.50/1M$15.00/1M272K
openai/gpt-5.4-miniOpenAI$0.75/1M$4.50/1M400K
anthropic/claude-sonnet-4.6Anthropic$3.00/1M$15.00/1M200K
google/gemini-3.1-proGoogle$2.00/1M$12.00/1M200K
x-ai/grok-4xAI$3.00/1M$15.00/1M256K

Model Selection

Pass the model ID in your request:

python
# Sarvam LLM
response = client.chat.completions.create(
    model="sarvam-30b",
    messages=[{"role": "user", "content": "Hello in Hindi"}]
)

# OpenRouter model
response = client.chat.completions.create(
    model="openai/gpt-5.4-mini",
    messages=[{"role": "user", "content": "Hello"}]
)

The API automatically routes to the correct provider based on the model ID prefix:

  • No prefix or sarvam/ β†’ Sarvam AI
  • openai/, anthropic/, google/, etc. β†’ OpenRouter
Was this page helpful?