OpenGPT-X: Teuken 7B

»Teuken 7B« is available in the following versions:

• »Teuken 7B-instruct-research-v0.4« for research purposes.
• »Teuken 7B-instruct-commercial-v0.4« for companies for commercial purposes under the »Apache 2.0« license. The model has already been optimized for chat through Instruction Tuning. »Teuken 7B-instruct-commercial-v.04« is comparable to the research version in terms of performance, although the research version achieves better results in the benchmarks by one to two percent.
  
• Teuken 7B-instruct-v0.6« and »Teuken 7B-base-v0.6« for non-commercial purposes under the license »CC BY-NC 4.0«. The update features significant improvements compared to »Teuken 7B-instruct-v.04«, including increased performance, improved robustness and reliability, and expanded application flexibility.
  

»Teuken 7B« is available for download free of charge and open source on Hugging Face.

Companies in particular have the opportunity to take part in a free webinar in which Fraunhofer scientists explain which applications can be realized with appropriate further processing on the basis of »Teuken 7B«.

»Teuken 7B-instruct-commercial-v0.4« is multilingual and has been optimized for chat through Instruction Tuning, so it can be used as a multilingual chatbot, e.g. in international customer service or to make company knowledge accessible to employees.

The following other applications can be implemented with »Teuken 7B«:

Areas of application:

• Summarize documents
  
• Generate texts
  
• Extracting information from texts
  

»Teuken 7B« can be further processed through Continued Pretraining, Finetuning, Instruction Tuning, Model Merging etc. in order to adapt the model to the company's own purposes. The result is a model that is optimized for the individual use cases in the company.

Select the version »Teuken 7B-instruct-commercial-v0.4«.

In order to adapt the model to your own business purposes, »Teuken 7B-instruct-commercial« can also be further processed with your own data through Continued Pretraining, Finetuning, Instruction Tuning, Model Merging, etc.

The model performs well in a performance comparison with other open source models, but still has potential for development in the areas of logical thinking, coding and mathematics. In addition, »Teuken 7B«, like other large language models, can generate content that is inappropriate, offensive or harmful.

»Teuken 7B-instruct« is a chatbot that is primarily intended for corporate applications and research projects. Developers from companies and the scientific community can use it to develop their individual chat applications. »Teuken 7B-instruct-commercial-v0.4« can also be further processed with your own data through continued pretraining, fine-tuning, instruction tuning, model merging, etc. to adapt the model to your own business purposes.

Yes, companies can use »Teuken 7B-instruct-commercial-v0.4« commercially for their AI applications under the »Apache 2.0 license«.

The basic model in version »Teuken 7B-base-v0.6« can be downloaded here for research purposes and for private, educational, and non-commercial use.

Basic models are particularly susceptible to the generation of inappropriate, offensive or harmful content. At the same time, base models offer the advantage that they can be developed into powerful special models through fine-tuning and instruction tuning if used correctly and responsibly.

Currently no. The EU AI Act will not apply until August 2025. AI models that were placed on the market before this date do not have to comply with the requirements of the EU AI Act until August 2027 (grandfathering).

The OpenGPT-X research project has been completed.

As this is an open source project, we also assume that adapted or specialized versions of the model will be developed for different applications by the scientific community or companies.

All evaluation results are available on our European Leaderboard:

• to the Leaderboard

For technical information on the model and its application, please refer to our model card:

• to the model card

We have summarized the key findings of the evaluation:

• to the benchmark graphics

A detailed review of the model can be found here:

• to the OpenGPT-X project website
  

Our scientists are in contact with the LLM community via the OpenGPT-X Discord server. This is also the place for questions and feedback about the model.

• join the discussion