Version 1.70

Article summary

Did you find this summary helpful?

Thank you for your feedback

Release Date: 30.10.2023

AI Module Note

Please use latests stable AI modules in each version update!

Tenant-Based Resource Management: Users can now customize the CPU/RAM resources for each tenant with a new setting called “Activate Premium Resources”. It allows users to speed up the train/inference times by allocating more resources to their specific needs. This setting only affects new models and not the ones that are already published.
Handling of Concurrent Training Requests: An error message will be displayed when users try to train a model as another model with the same model ID is already being trained. The error message will say: “The training is still in progress for the same model ID: modelId”. This will help users avoid potential conflicts and manage concurrent training requests better.

Restriction on Tenant Name: Tenant name can now only contain lowercase letters (a-z), numbers (0-9), and hyphens ("-"). Hypens ("-") are not allowed at the beginning and end of the text. Tenants created in prior versions will remain unaffected by this change.
Bug Fix: An issue that prevented the system from creating the correct number of replicas for expired inference pods after training with modified train data has been fixed. Now, the system ensures that the number of replicas matches the original setting.

Was this article helpful?

Table of contents