Ground3D-LMM-Reason3D-4B-Joint

Fine-tuned Ground3D-LMM checkpoint for the Reason3D benchmark (Table 5), 3D+2D (point cloud + 20 RGB views) setting.

Checkpoint file: pytorch_model.pth
Eval config: configs/ground3dlmm_eval_reason3d_image.py
Code + data setup: https://github.com/AmolHarsh/ground3d-lmm (see docs/DATA_SCANREFER_REASON3D.md)

Usage

CKPT=$(huggingface-cli download amolharsh/Ground3D-LMM-Reason3D-4B-Joint pytorch_model.pth)
bash tools/dist_test.sh configs/ground3dlmm_eval_reason3d_image.py "$CKPT" 4

Downloads last month: -; Downloads are not tracked for this model. How to track

Inference Providers NEW

This model isn't deployed by any Inference Provider. 🙋 Ask for provider support