TextSynth provides access to large language models thru a REST API and a playground. They can be used for example for text generation, question answering, classification, chat, ...

TextSynth employs a custom inference framework based on LibNC to get fast inference on standard GPUs and CPUs.

The basic service is free and offers several requests per minute even for large models such as GPT-J. In order to access the API or to get the best performance with the playground, you need to create an account. If you wish a higher bandwidth or custom support, please contact us: contact at textsynth dot com.