Ostris AI Toolkit Adds LoRA Training for MiniMax H3
The toolkit now supports LoRA training for MiniMax H3, initially limited to text-to-video and image-to-video. The announcement says it uses quantized Comfy weights.
Ostris AI Toolkit has added support for training LoRAs for MiniMax H3. According to the announcement, the feature uses quantized Comfy weights.
For now, support is limited to text-to-video and image-to-video tasks. Engineers evaluating the toolkit for video-model fine-tuning should take this scope into account.
Before testing, check the project’s documentation and current limitations, and validate the workflow with test data. The announcement does not describe other training modes or performance results.
If you use AI to study or apply this material, share only the necessary excerpts and avoid putting personal data or confidential organizational content in prompts.