Apertus is a large-scale open, multilingual language model developed by EPFL, ETH Zurich, and the Swiss National Supercomputing Centre (CSCS). It is designed to be transparent and diverse, trained on 15 trillion tokens across more than 1,000 languages. Apertus serves as a building block for developers and organizations for future applications such as chatbots, translation systems, or educational tools.
The overall score is a composite metric that summarizes the content quality across all evaluated dimensions.
Overall Score
7.9
Score
0.0
Median
Loading...
The content flows well from one idea to the next. It should not be a heap of related information, but rather should have continuity between paragraphs.
Fluency
9.0
Score
0.0
Median
Loading...
The content includes keywords that are likely to appear in LLM prompts.
Keywords
9.0
Score
0.0
Median
Loading...
For the content to be authoritative, it must consistently demonstrate persuasion, credibility, evidence based claims.
Authority
9.0
Score
0.0
Median
Loading...
The content references credible sources, studies, or expert opinions to back up its claims.
Citations
4.0
Score
0.0
Median
Loading...
The writing is simple and can be easily understood.
Simplicity
8.0
Score
0.0
Median
Loading...
The content includes quantitative statistics instead of qualitative discussion, wherever possible.
Statistics
8.0
Score
0.0
Median
Loading...
There is the inclusion of technical terms and uncommon words within the text rather than generic words and phraseology.