https://docs.ai.gwdg.de/en/categories/service/index.htmlHugoenhttps://docs.ai.gwdg.de/en/user/basics/request/index.htmlMon, 01 Jan 0001 00:00:00 +0000https://docs.ai.gwdg.de/en/user/basics/request/index.htmlHow a request is checked Every request passes three gates before it reaches a model: the API key it was made with, the user account that key belongs to, and the organisation that account belongs to. Each gate holds its own set of limits, split into the two kinds above: a request rate, counted in requests, and a budget, counted in euros. The request is measured against every limit that is set at that level, and it is rejected the moment any one of them is out of allowance, without the later gates being consulted. Only when all three gates pass does the request reach a model, and its cost is then written back to the budgets of all three levels at once.https://docs.ai.gwdg.de/en/user/ai-services/arcana/rag-service/index.htmlMon, 01 Jan 0001 00:00:00 +0000https://docs.ai.gwdg.de/en/user/ai-services/arcana/rag-service/index.htmlWarning This service is currently in beta phase and is updated regularly. The same applies to the documentation. RAG (Retrieval-Augmented Generation) is an advanced AI technique designed to improve the accuracy, reliability, and contextual relevance of AI-generated responses. Traditional AI models, such as large language models (LLMs), rely solely on pre-trained data to generate answers. While these models can provide insightful responses, they are limited by the information they were trained on, which may become outdated or may not cover specific topics in detail.https://docs.ai.gwdg.de/en/user/ai-services/arcana/getting-started/index.htmlMon, 01 Jan 0001 00:00:00 +0000https://docs.ai.gwdg.de/en/user/ai-services/arcana/getting-started/index.htmlWarning This service is currently in beta phase and is updated regularly. The same applies to the documentation. Table of contents: First Login Creating an Arcana Uploading Files Viewing your Access Link Updating files in an Arcana The process screenshots contain blank blocks. These are in the positions that will be replaced with the username. First Login Go to the main Arcana page and click on register. You will need to log in with your Academic Cloud account first.https://docs.ai.gwdg.de/en/user/ai-services/arcana/how-to-use/index.htmlMon, 01 Jan 0001 00:00:00 +0000https://docs.ai.gwdg.de/en/user/ai-services/arcana/how-to-use/index.htmlWarning This service is currently in beta phase and is updated regularly. This also applies to the documentation. Warning Previously, RAG/Arcanas only worked with specific models. This is no longer the case, and Arcanas now work with every model. The previous method of using Arcanas with models that have ‘RAG’ in their name (or the little book icon) will not be supported in future. Please use the GWDG Tools as described below!https://docs.ai.gwdg.de/en/procurement/index.htmlMon, 01 Jan 0001 00:00:00 +0000https://docs.ai.gwdg.de/en/procurement/index.htmlAvailable AI Products Within the Academic Cloud The Academic Cloud is a modular service portal available nationwide for universities, colleges, and research institutions. As a shared platform, it enables the provision, development, and use of digital services across institutional boundaries and supports cross-institutional collaboration.https://docs.ai.gwdg.de/en/administration/index.htmlMon, 01 Jan 0001 00:00:00 +0000https://docs.ai.gwdg.de/en/administration/index.htmlInfo We will shortly be updating the documentation to include guidance on budget and API key management for organizational administrators.https://docs.ai.gwdg.de/en/user/index.htmlMon, 01 Jan 0001 00:00:00 +0000https://docs.ai.gwdg.de/en/user/index.htmlInfo Soon you will find the user documentation of our AI services here. It will contain detailed information about the individual AI services, AI models, web interfaces, terms of use, data protection, and more. As part of the revision of this documentation, we will make various changes to the organization of the content.https://docs.ai.gwdg.de/en/technical/saia-platform/index.htmlMon, 01 Jan 0001 00:00:00 +0000https://docs.ai.gwdg.de/en/technical/saia-platform/index.htmlSAIA is the Scalable Artificial Intelligence (AI) Accelerator that hosts our AI services. Such services include Chat AI and CoCo AI, with more to be added soon. SAIA API (application programming interface) keys can be requested and used to access the services from within your code. API keys are not necessary to use the Chat AI web interface. The SAIA API is suitable for interactive inference scenarios. If you have a large amount (eg. thousands of LLM queries) of requests that you can process asynchronously, the batch paradigm of our HPC cluster is the better choice. Your batch will be completed more predictably, in less time, and with lower cost. Check out how to get started with our HPC cluster and then running LLMs to learn how you can setup up a batch inference job on the cluster. vLLM is another popular choice for LLM inference.https://docs.ai.gwdg.de/en/procurement/chatai_free/index.htmlMon, 01 Jan 0001 00:00:00 +0000https://docs.ai.gwdg.de/en/procurement/chatai_free/index.htmlChat AI Free Chat AI Free provides access to the open-weight models hosted by GWDG in Göttingen for testing and evaluation purposes. All users must have an Academic ID and a valid email address. The service is free of charge and can be accessed on an individual basis. Limitations for Chat AI Free Minimal cluster resources are allocated for the free variant. Reduced list of available open-weight AI models. Access only to Chat AI service, Arcana/RAG, and CoCo AI. No access to Image AI, Voice AI, or Protein AI. Only one API key per user. Provided on a best-effort basis: No guaranteed availability. No guaranteed response times. Reduced request rate limits. Contractual Terms There is no additional contract required for “Chat AI Free.” Users must accept our Terms of Use by agreeing to the terms provided in the links above.https://docs.ai.gwdg.de/en/procurement/chatai_prepaid/index.htmlMon, 01 Jan 0001 00:00:00 +0000https://docs.ai.gwdg.de/en/procurement/chatai_prepaid/index.htmlChat AI Prepaid Chat AI Prepaid provides access to the open-weight models hosted by GWDG in Göttingen. It is designed for organizations that desire fixed budgets and annual payment plans while maintaining some budget flexibility during the ongoing contract cycle. The contract is valid for 12 months and can be extended upon request. The organization must also sign an Academic Cloud Basic contract. Using the AI models via our web interface or API is optional.https://docs.ai.gwdg.de/en/procurement/chatai_flexible/index.htmlMon, 01 Jan 0001 00:00:00 +0000https://docs.ai.gwdg.de/en/procurement/chatai_flexible/index.htmlChat AI Flexible Chat AI Flexible provides access to the open-weight models hosted by GWDG in Göttingen. It is designed for organizations that require monthly billing plans for flexible usage of AI models. Typical use cases include publicly accessible applications. The contract is valid for 12 months and can be extended upon request. The organization must also sign an Academic Cloud Basic contract. Using the AI models via our web interface or API is optional.https://docs.ai.gwdg.de/en/procurement/chatai_external_prepaid/index.htmlMon, 01 Jan 0001 00:00:00 +0000https://docs.ai.gwdg.de/en/procurement/chatai_external_prepaid/index.htmlChat AI External Models Prepaid Chat AI External Models Prepaid provides access to the externally hosted AI models. It is designed for organizations that desire fixed budgets and annual payment plans while maintaining some budget flexibility during the ongoing contract cycle. The contract is valid for 12 months and can be extended upon request. The organization must also sign an Academic Cloud Basic contract. Using the AI models via our web interface or API is optional.https://docs.ai.gwdg.de/en/procurement/chatai_external_flexible/index.htmlMon, 01 Jan 0001 00:00:00 +0000https://docs.ai.gwdg.de/en/procurement/chatai_external_flexible/index.htmlChat AI External Models Flexible Chat AI External Models Flexible provides access to externally hosted AI models. It is designed for organizations that require monthly billing plans for flexible usage of AI models. Typical use cases include publicly accessible applications and internal tools with variable usage patterns. The contract is valid for 12 months and can be extended upon request. The organization must also sign an Academic Cloud Basic contract. Using the AI models via our web interface or API is optional.https://docs.ai.gwdg.de/en/technical/saia-platform/limits_budget/index.htmlMon, 01 Jan 0001 00:00:00 +0000https://docs.ai.gwdg.de/en/technical/saia-platform/limits_budget/index.htmlFrom a contract to a budget An organisation’s budgets come from what it has bought: Step Calculation Example Annual budget by contract 72,000 € Monthly budget annual ÷ 12 6,000 € Daily budget monthly ÷ 15 400 € The daily figure is divided by 15, not 30, and that is deliberate. At ÷ 30 you could never use more than an even day’s share, and a deadline or a teaching week would hit the wall. Dividing by 15 gives roughly twice the even daily rate as burst allowance: heavy days are possible, and sustaining that pace all month is not.https://docs.ai.gwdg.de/en/technical/saia-platform/limits_rate/index.htmlMon, 01 Jan 0001 00:00:00 +0000https://docs.ai.gwdg.de/en/technical/saia-platform/limits_rate/index.htmlHow did we calculate rate and budget limits for each organization? Two separate questions, with different answers. An organisation’s limits are derived from its contract, so that what it bought and what it may spend per day agree. The unit that connects euros to requests A budget is measured in euros and a rate limit in requests, so translating one into the other needs a conversion factor. Derived from real usage statistics at GWDG: An average request consist of roughly 10,000 tokens, which at around 1 € per million tokens comes to about 0.01 € per request.https://docs.ai.gwdg.de/en/user/ai-services/arcana/api-usage/index.htmlMon, 01 Jan 0001 00:00:00 +0000https://docs.ai.gwdg.de/en/user/ai-services/arcana/api-usage/index.htmlWarning This service is currently in beta phase and is updated regularly. The same applies to the documentation. The Arcana Manager exposes and api endpoint for you to use. With the api endpoint you can perform all the actions that the frontend Arcana Manager is capable of. To use the API, you’ll first need an API key, which you can request on our website. This key works across all our services. If you’re already using the LLMs via the api, you can use the same key. Otherwise you can request your key following this documentation.https://docs.ai.gwdg.de/en/user/ai-services/arcana/index.htmlMon, 01 Jan 0001 00:00:00 +0000https://docs.ai.gwdg.de/en/user/ai-services/arcana/index.htmlArcana is a Retrieval-Augmented Generation (RAG) service that enables you to interact with Documents - such as research papers, manuals, or study materials - using natural language. The Arcana service works together with our Chat AI Service. LLMs benefit from RAG (Retrieval-Augmented Generation) by using the relevant information provided in a Document Collection (Arcana) for generating more accurate, reliable, and contextually grounded responses. In order to use it, you need to activate your account for the Arcana page and set up an Arcana following the Getting Started Guide. Once this is done, you can share it with colleagues or the public. Details about this process can be found in the usage examples.https://docs.ai.gwdg.de/en/user/ai-services/arcana/docling-process/index.htmlMon, 01 Jan 0001 00:00:00 +0000https://docs.ai.gwdg.de/en/user/ai-services/arcana/docling-process/index.htmlWarning This service is currently in beta phase and is updated regularly. The same applies to the documentation. The RAG Service provides users with an efficient way to upload and process PDF documents using Docling. The system converts uploaded PDFs into Markdown format while also automatically annotating them. This enhanced Markdown output, referred to as Markdown Plus, includes metadata and structural annotations for improved document parsing and customization.https://docs.ai.gwdg.de/en/user/ai-services/arcana/public-arcanas/index.htmlMon, 01 Jan 0001 00:00:00 +0000https://docs.ai.gwdg.de/en/user/ai-services/arcana/public-arcanas/index.htmlWarning This service is currently in beta phase and is updated regularly. The same applies to the documentation. Here is a list of public links for Arcanas. These are sorted in categories and contain public material. Please refer to the How to use section of you need help working with these. GWDG Services Contains knowledge about the GWDG Services and Documentation. Example question: Welche Dienste könnte ich als Forscher bei der GWDG nutzen? Institute for Computer Science Contain knowledge about the computer science study track. Example question: Wie funktioniert der Studiengang angewandte Informatik? In case you find a broken link, please let us know via our support contacts.https://docs.ai.gwdg.de/en/user/ai-services/arcana/support/index.htmlMon, 01 Jan 0001 00:00:00 +0000https://docs.ai.gwdg.de/en/user/ai-services/arcana/support/index.htmlWarning This service is currently in beta phase and is updated regularly. The same applies to the documentation. If you run into problem using this service please contact our KISSKI support We also need your help providing publicly available Arcana links. If you have one that you think is relevant to a larger group of users, please also reach out to us so we can publish it.