Production-ready voice agents built on the best voice models.

Turn any text agent into a world-class conversational agent, deployed anywhere.

Contact Sales Try Cartesia

async def get_agent(env, call_request):

return LlmAgent(

model="gpt-5.4", # Or "gemini/gemini-3.5-flash", etc.

tools=[web_search],

config=LlmConfig(

system_prompt="You are a helpful assistant.",

introduction="Hello! How can I help you today?",

)

app = VoiceAgentApp(get_agent=get_agent)

if __name__ == "__main__":

app.run()

The smoothest developer experience, with full flexibility.

Agents scale from zero to enterprise scale with a simple developer experience, built-in evals, and enterprise features like SIP trunking.

Robust SDK

Our SDK is everything you need to build the most powerful, intelligent agents with taste. Get advanced reasoning, parallel background tasks, and real-time system actions —all with ultra-low latency.

Top notch in-house models

Our fully-owned voice stack is underpinned by our leading Ink STT and Sonic TTS models plus deployment so you can fine-tune, co-locate, and optimize for max performance.