New Model (GPT-5.2), Incident Integration, and Improved Document Search
- New Language Model:
- Ritchy now uses GPT-5.2 by OpenAI as its underlying language model. Compared to the previous GPT-4.1 model, it provides improved reasoning capabilities and a deeper understanding of language.
- Automatic Integration of Incident Reports:
- Ritchy can now automatically detect whether a user request relates to a current incident or maintenance activity of the IT Center. If a match is found, the corresponding incident report is directly incorporated into the response, including status information and a link to the incident page. Incidents are prioritized by severity (incident > partial incident > maintenance > warning > notice).
- Improved Document Search:
- The way Ritchy accesses the IT Center documentation has been fundamentally redesigned. The search now uses a three-stage approach (keyword search, vector search, and semantic re-ranking) to retrieve more relevant articles. In addition, more document context is included in responses, resulting in more precise and complete answers.
- More Reliable Language Detection:
- Automatic language detection has been made more robust. In particular, for queries containing English technical terms (e.g., product names like Cisco AnyConnect), the system now more reliably identifies the user’s actual language and consistently formulates the response in that language.
- Structural Improvements:
- The internal architecture of the application has been modularized to simplify future extensions and maintenance. These changes do not result in any visible differences for end users.
Improved Model Selection & Overview (v1.16.0)
Bugfixes and Improvements:
- The model selection list is now scrollable when it becomes too long
- The model overview has been redesigned for improved clarity
New Models Available for Employees and Students (v1.16.0)
Features:
- For employees, the following new models have been added::
- Mistral-4-small (für API-Zugriff freigeschaltet)
- GPT-5.3 (Chat)
- GPT-5.4-mini (Chat) (für API-Zugriff freigeschaltet)
- GPT-5.2 (Reasoning) (für API-Zugriff freigeschaltet)
- The following models have been removed:
- gpt-5.1
- gpt-5.0
- gpt-4.1
- Llama-3.3-70B-Instruct
- mistral-small-3.2-24B-instruct-2506
- For students, the following new models have been added:
- Mistral-4-small (für API-Zugriff freigeschaltet)
- GPT-5.4-mini (Chat)
- GPT-5.2 (Reasoning)
- The following models have been removed:
- gpt-5.1
- gpt-5.0
- gpt-4.1
- Llama-3.3-70B-Instruct
Bugfixes and Improvements:
- The model selection list is now scrollable when it becomes too long
- The model overview has been redesigned for improved clarity
Opencast: Upgrade to version 19
- various bugfixes
- Paella Player 8
- all Release Notes
Release Notes Version 2.50
Improvements and bug fixes:
- The display of study room opening hours has been improved
- Various bug fixes
Adjustment of API Endpoints
For everyone using our API: Some endpoints have been revised and are now more clearly structured, functionally expanded, and generally easier to use. Specifically, this means that interactions with a resource’s metadata and graphs will now take place via Graph, GraphMetadata and GraphMetadataContent. Access to objects and files can now be found under Storage and StorageContent. If you use these functions in your scripts, they must be updated promptly. The previous endpoints are already marked as deprecated in the Swagger documentation and are expected to be disabled in August. We will inform you of this again in advance via the newsletter and through relevant notices on the platform.
Information for Reviewers of Coscine Storage Space Requests
Important information for reviewers of storage space requests: The “Restrictedly suitable” option is no longer available as a final review result. As of March 16, 2026, reviewers must now choose between ‘suitable’ and “not suitable.” RDM staff and 1st-level support have already been informed of this change during the monthly consultation hour. If you are part of the RDM staff but have not yet received an invitation to the monthly consultation hour, please feel free to contact us via e-mail. You are also welcome to check out the support services.
Support for New Models (GPT-5.3-Chat, Mistral Small 4) (v1.15.0)
Features:
- Added support for new models:
- gpt-5.3-chat, Mistral-Small-4-119B-2603
Bugfixes and Improvements:
- No spaces or special characters are allowed anymore in internal deployment names
- Improved performance when accessing the API
- For image generation, the input image is now only displayed if it was manually selected
- Minor improvements to shared chats, especially regarding model selection and chats with images
- Improved behavior when manually cancelling a completion
- Fixed an issue with title generation when it takes longer than the actual prompt
- Fixed a problem with database indexes that prevented the app from starting
Support for New Models (GPT-5.3-Chat, Mistral Small 4) (v1.15.0)
Features:
- Added support for new models:
- gpt-5.3-chat, Mistral-Small-4-119B-2603
Bugfixes and Improvements:
- No spaces or special characters are allowed anymore in internal deployment names
- Improved performance when accessing the API
- For image generation, the input image is now only displayed if it was manually selected
- Minor improvements to shared chats, especially regarding model selection and chats with images
- Improved behavior when manually cancelling a completion
- Fixed an issue with title generation when it takes longer than the actual prompt
- Fixed a problem with database indexes that prevented the app from starting
Default Model Update and gpt-40 and gpt-40-mini Model Discontinuation (v1.14.0)
Note:
- The models gpt-4o and gpt-4o-mini will be discontinued on March 31, 2026, and removed from the model list. They will also no longer be available when using an API key after this date. To continue existing chats, they must be migrated to another model.
Features:
- gpt-oss-120b is now set as the default model
Bugfixes and Improvements:
- Title generation is no longer counted toward the maximum number of parallel requests per user
- Added support for the
encoding_formatparameter in the embedding endpoint - Restricted access to own chats in user groups that are no longer accessible
