TextSynth provides access to large language or text-to-image models such as Mistral, Llama, Stable Diffusion, Whisper thru a REST API and a playground. They can be used for example for text completion, question answering, classification, chat, translation, image generation, speech to text transcription, ...

TextSynth employs custom inference code to get faster inference (hence lower costs) on standard GPUs and CPUs. The site was founded in 2020 and was among the first to give access to the GPT-2 language model.

All our servers are located in France.

The basic service is free but rate limited. Users wishing no limitation can pay a small amount per request (see our pricing). If you wish custom support, please contact us: contact at textsynth dot com.