| license: mit | |
| base_model: | |
| - luodian/OTTER-MPT7B-Init | |
| pipeline_tag: image-to-text | |
| The visual instruction-tuned model for Nature Communications paper **"Towards a holistic framework for multimodal LLM in 3D brain CT radiology report generation"** (doi.org/10.1038/s41467-025-57426-0) |