FAQ - RWTHgpt

• GPT (Generative pre-trained transformers) describe AI models that are able to create human-like texts and content (images, music and more) with the help of previously trained large language models.

• In the context of RWTHgpt, tokens are the unit with which the AI models used measure the length of texts (both for user-generated questions and system-side answers).
  These are frequent character strings that occur in a text. The division into tokens is complicated and is not uniformly based on words or characters.
• Detailed information on calculating the number of tokens and the background can be found here:
  • OpenAI GPT Token Guide (DE)
  • What are Tokens
• OpenAI itself offers a calculator that makes it possible to determine the number of tokens and calculation in sentences or texts:
  • Tokenizer
    Please note that this page is subject to the general data protection guidelines of OpenAI. Please do not insert any personal data (from third parties) here.

• There is a general token limit for chats with RWTHgpt, which can vary depending on the selected language model or AI tool.
• The maximum number of tokens that can be used, as well as the number of tokens used so far, is displayed below the text input field.
• As soon as the token limit is reached, either open a new chat or edit prompts.
• Individual entries (text & image) have no additional token limit.
• Depending on the selected language model or AI tool, there are also restrictions on the number of prompts per hour. These can be seen when selecting the model.

• Prompt is a term for a text input that is passed on to the AI.
• Both user input and system-side instructions can be described as a prompt.

• Detailed descriptions and notes can be found here:
  • Explanation of prompts and prompt engineering (DE)
  • OpenAI Prompt Engineering Guide

RWTHgpt can recognize and analyze images.
However, the correct model must be used, as only the OpenAI GPT 4o and OpenAI o1 models have this capability.

Unfortunately, it is not yet possible to generate images.

• The two RWTHgpt language models OpenAI GPT 4o and OpenAI o1 can recognize and analyze images.
• Image generation is not yet possible, but will be made available in Q4 2025.

• Click on the paperclip icon to the left of the chat
• Select the image to be analyzed
• Now write a prompt that instructs RWTHgpt to analyze the image. You can also set restrictions here.
  • For example: “Analyze the attached image (focusing on the font and ignoring colors).