Multi Provider LLM Proxy for Voice Agents

An OpenAI-compatible chat completions proxy that enables voice agents to use multiple LLM providers including Amazon Bedrock Agents and OpenAI. Provides an extensible architecture for switching between providers through environment variables or request parameters, supporting both streaming and non-streaming responses with exact OpenAI format matching.

No demo yetView source Fork template

The numbers

latency—

cost / min—

frameworkCustom

The stack

telephonyWeb Only

speech-to-textDeepgram Nova-3

llmGPT-4o-mini

text-to-speechDeepgram Aura-2

System prompt

No prompt published.

Config

config.json

{
  "default_model": "gpt-4o-mini",
  "sse_streaming": true,
  "provider_options": [
    "bedrock",
    "openai"
  ],
  "openai_compatible": true,
  "supports_streaming": true
}

Voice AI recipes, picks, and analysis.

Get the useful new templates plus the occasional teardown of what’s working in production voice AI.