TextSynth

Through a web playground and a REST API, TextSynth gives you access to open language models, image generation, speech-to-text and text-to-speech.

It is an AI API service that covers text completion, chat, translation, transcription and embeddings from one account.

Models include Mistral 7B, Llama3 8B, Llama3.3 70B instruct, Mixtral 47B instruct, GPT-J 6B and MADLAD400 7B for translation, plus Stable Diffusion for images, Whisper Large V3 for transcription and Parler-TTS for speech.

The free plan lets you try every language model, but with rate limits, regular captcha checks, a cap of 200 generated tokens per request and speech transcription limited to 3 minutes.

All free requests are logged on the servers.

The Standard plan removes the rate limits and token cap, and request content is not logged.

It is credit based, so you cannot get a surprise bill.

Prices run from $0.20 per million input tokens on the smaller models to $0.70 on Llama3.3 70B, and Stable Diffusion costs $0.005 per image.