The best open source AI on demand in a sovereign cloud

Optimised to handle large amounts of text ensuring consistency across multiple sources

Excellent in development, programming and academic research tasks

High multilingual flexibility with more than 30 languages supported

Suitable for artists and content creation, including storytelling

Larger training corpus than Mixtral 8x7B for more complex tasks

Able to analyse unstructured data to support decision making and generate content

Management of conversational subtleties to feed complex discussions

Optimised for logical exploration (combining complex information) and generating ideas (scenarios, etc.)

Economical and very fast for many common tasks

Ideal for summarising, moderating, calculating, coding and extracting data from unstructured sources

Suitable for real-time interpretation of data and for logical reasoning

Easy to adjust and contextualise in order to limit undesirable outcomes

The most powerful open source embedding model on the market

The reference for semantic search and augmented search (ASR) tasks

Ideal for advanced use of embedding vectors in a variety of applications

Outstanding performance, whatever the language of the text (100 languages)

This model is the result of community work based on a model published by Microsoft.

Excellent value for money, ideal for prototyping and simple tasks with limited resources

Interesting performance for relatively simple tasks, whatever the language of the text

Extreme speed for indexing huge databases or real-time processing

High energy efficiency to reduce environmental impact

Model trained on over 1 million hours of data

Transcription errors reduced by up to 20% compared with Whisper V2

Better handling of accents, background noise and complex speech (e.g., calls or videoconferences)

Enhanced multilingual support and translation of transcriptions into languages other than English

Audio transcription in over 57 languages and translation of transcribed text into English

Model trained on 680,000 hours of data in 98 languages

Automatic identification of the original language

The best combination of quality and speed in generative AI image creation

Fast generation of photo-realistic images in 1, 2, 4 or 8 steps based on prompts

Operates by distillation, which increases energy efficiency and ensures excellent quality

Optimised for English, with limited knowledge of other languages (FR, DE, ES, IT, etc.)

Create photos in multiple styles from one or more profile photos

Powerful and flexible: recontextualisation, colourisation, age and gender change, mix of identities, etc.

Outstanding image quality, surpassing DALL-E 3 and MidJourney in some areas

Prompt fidelity and precise interpretation of complex scenes

A wide range of styles