Skip to main content

What is Luminous?

Luminous Model Family

The Luminous series is a family of large language models. Large language models are powerful technological tools that can process and produce human text. They have acquired this ability by “reading” enormous amounts of human text data during training. Similar to someone reading a whole library and half of the internet, large language models have acquired a structural understanding of language and accumulated information about the world.

The Luminous family currently consists of three models, which differ in complexity and ability. They are, from the smallest to the largest, Luminous-base, Luminous-extended and Luminous-supreme. All Luminous models have been trained in the five most commonly spoken European languages: English, German, French, Italian and Spanish. Contrary to other large language models, the Luminous family posseses the ability to work with images as well. We call this multimodality. The Luminous family consists of three available models with varying parameter sizes and capabilities.

Luminous-supremeLuminous-supreme is the largest model in the Luminous model family and consequently the most capable one. It can solve all tasks the smaller models can solve as well and is especially well suited for creative text writing.
Luminous-supreme-controlLuminous-supreme-control is based on Luminous-supreme and is optimized on a certain set of tasks such as Q&A and Natural Language Inference. One huge advantage of Luminous-supreme-control is, that it has a much better zero-shot performance.
Luminous-extendedLuminous-extended is our second-largest model. It is well suited for tasks like information extraction and language simplification. It performs better on a wide range of tasks compared to Luminous-base and is faster and cheaper than Luminous-supreme.
Luminous-baseLuminous-base is our smallest model. That makes it the fastest and cheapest model in the Luminous model family. Therefore, this model is suited for applications where speed is important and costs should be low. It is very well suited for tasks like classification and labelling.

Model Capabilities, Methods and Services

All of our models can handle text as input. Next to that, Luminous-base and Luminous-extended can also handle multimodal input. This means they can process a combination of image and text input and output text. Read more about it here.

A method abstracts a specific interaction with a model for a low-level purpose. Luminous-Explore is our model for semantic embeddings based on Luminous-base.

A service provides value on top of a model or independent of a specific model. These services are designed for a certain task, such as question answering or summarization.

Interacting with Luminous models

You can interact with a Luminous model by sending it a text. We call this a prompt. It will then produce text that continues your input and return it to you. This is what we call a completion. By using a set of techniques, which we lay out in the “Prompting” section, you can instruct our models to solve a wide variety of text-based tasks. Note that increasing task complexity may require larger models. Find examples for tasks of varying complexities below:

Continue the proverb: An apple a day keeps the doctor away.
Find a suitable topic for the following text: On the day we went public, our shares rose 23%.
Topic: Finance
Summarize the following news article:
Lumi provides support, for example, when dealing with the authorities and knows how to re-register one's residence within Heidelberg or when the paper garbage will be collected. To this end, the AI draws on publicly available information from the city of Heidelberg to provide the most tailored answer possible for each request. In order for Lumi to learn what exactly Heidelberg residents are interested in and how they formulate their concerns, the AI has to gather experience - and that works best in conversation with users.
Summary: Lumi is an AI that helps people navigate the bureaucracy of Heidelberg. It learns by talking to people.

Other Tasks

Apart from generating completions, our luminous models possess the ability to semantically compare texts, summarize documents, perform Q&A tasks and more.

To find out how you can perform these tasks with our Python API, please refer to the Tasks section.


All models provided in the API were trained in five languages (English, Spanish, German, French, Italian), so you can prompt and work with the model in every one of these languages.

Ein blindes Huhn findet auch mal ein Korn.

German equivalent to: “A blind man may sometimes hit the mark.”