Pass IndexThe State of AISign in

csm-1b

CSM (Conversational Speech Model) is a speech generation model from Sesame that generates RVQ audio codes from text and audio inputs.

text → audio · made by sesame

$0.000007per character
DeepInfra

Sold by 1 way

SellerLaneRate
DeepInfra aggregatorstandard$7per million charactersapi.deepinfra.com · read 2026-08-25

About

csm-1b — an audio model from sesame, sold by one company from $7 per million characters.

It takes text and returns audio. The catalogue files it under speak. Only DeepInfra sells it, at $7 per million characters.

Every current figure

Maker
sesame
Register
model
Takes
text
Returns
audio
Parameters
1 billion · read from its own name
Licence
not read
Sellers
1
Maker's own price
not read
Price
$7 per million characters — DeepInfra

Known as 1 name

sesame/csm-1b