Simple and flexible pricing
Pay only for what you use, with no commitment period, and keep your data confidential with the best open-source AI.
Start with 1M free creditsInteract, analyse and create content with AI
mistralai/Ministral-3-14B-Instruct-2512
Input tokens:
Output tokens:
Apertus-70B-Instruct-2509
Input tokens:
Output tokens:
Qwen/Qwen3.5-122B-A10B-FP8
Input tokens:
Output tokens:
google/gemma-4-31B-it
Input tokens:
Output tokens:
moonshotai/Kimi-K2.6
Input tokens:
Output tokens:
nvidia/NVIDIA-Nemotron-3-Nano-30B-A3B-FP8
Input tokens:
Output tokens:
mistralai/Mistral-Small-4-119B-2603
Input tokens:
Output tokens:
Boost your searches with open source
BAAI/bge-reranker-v2-m3
Input tokens:
Qwen/Qwen3-Reranker-0.6B
Input tokens:
Turn your data into semantic vectors
Bge Multilingual Gemma2
Creation of the text embedding vector
All MiniLM L12 v2
Creation of the text embedding vector
Free
Qwen/Qwen3-Embedding-8B
Creation of the text embedding vector
Transcribe audio files into text with AI
Whisper V3
Speech recognition
Create images from text with AI
Flux schnell
Creation of images from text
Photomaker V2
Customisation of realistic human photos from images and text
Questions? Doubts?
Here we have gathered the questions we are asked most frequently.
At the moment, we do not offer the ability to train our AI Services with custom data.
Yes, AI Services can be integrated into your applications, which you are free to monetize.
At the moment, the ability to create or interpret images with AI Services is under review. We will offer this feature once it is fully operational and able to ensure total control of your data.
Yes, AI Services is a multi-model product that includes Whisper and lets you transcribe audio files into text.
The LLM models we offer have different limits, due to variations in model complexity and size. However, you can find the maximum prompt size accepted by our models here
Tokens are units that split text before the API analyzes it; they are characters or a set of words. In general, a token corresponds, for example, to 4 characters in English or to 3/4 of a word, depending on the language.
Input tokens refer to the questions asked to the AI and output tokens to the answers it generates.
No, the models we offer are based on a huge amount of data, which does not need to be continuously fed with new information.
- We use various Open Source models to offer a wide range of tools with AI Services, such as Mixtral for text creation and Whisper for audio transcription.
- You can find the full list of the models we offer here
You can use AI immediately and monitor your consumption in real time. The invoice will be issued at the end of the month based on the number of tokens consumed. When finalizing the creation of your account, we will ask you to confirm your identity by registering a valid payment card, thereby strengthening our anti-fraud measures.
Yes, the various models we offer are based on Open AI standards to make it easier to switch to our AI Services and to use it.
Transcription or image-creation models are billed per minute of creation. The time frame depends on various parameters, such as the size of the file to be transcribed (for transcription) or the image quality and the number of images created simultaneously (for images).
You are billed for the creation time used. For example, an image that takes 4 seconds to create with SDXL-Lightning will be billed in proportion to the stated rate, i.e. 0,02 €.