Are all the TTS/Speech models down?

It fails through API calls. The only combination I was able to use was model=“sesame/csm-1b”, voice=“alloy” and response_format=“mp3”. Are there any information on which voices and formats available for each model?