GLM-5.3 is live. Run Z.ai's latest model on Friendli Model APIs. Try it today ➜
JohnZhan
Available on FriendliAI
Run this model inference on single tenant GPU with unmatched speed and reliability at scale.
Run this model inference with full control and performance in your environment.
Model Details
Model Provider
JohnZhan
Model Tree
Qwen/Qwen3-VL-8B-Instruct
Input Modalities
Output Modalities
Supported Functionality
MotionInsight diagnoses object motion in generated videos using RGB frames, object tracks, and camera motion. It produces reasoning and three scores from 1 (poor) to 5 (excellent):
| Dimension | Output field |
|---|---|
| Object consistency | structural_stability |
| Motion continuity | motion_coherence |
| Physical plausibility | physical_plausibility |
The model is based on Qwen3-VL-8B-Instruct and uses four BF16 Safetensors shards (approximately 17.7 GB).
Use Python 3.10 and the custom Qwen3-VL patch from the code repository:
bash
git clone --recurse-submodules https://github.com/JohnZhan2023/MotionInsight.gitcd MotionInsightpython3.10 -m venv .venv-modelsource .venv-model/bin/activatepython -m pip install --upgrade pippython -m pip install huggingface-hub==0.36.2hf download JohnZhan/MotionInsight-8B --local-dir checkpoints/MotionInsight-8Bpython -m pip install torch==2.11.0 torchvision==0.26.0 \ --index-url https://download.pytorch.org/whl/cu126python -m pip install -r checkpoints/MotionInsight-8B/requirements/inference.txtpython scripts/patch_transformers.pyAfter extracting motion features, run:
bash
python inference.py \ --model_name_or_path checkpoints/MotionInsight-8B \ --video_path /path/to/video.mp4 \ --object_motion_path /path/to/object_motion.pt \ --camera_motion_path /path/to/camera_motion.pt \ --target "tennis ball" \ --output_path outputs/prediction.jsonlExample answer format:
text
<thinking>Diagnostic reasoning about the target object's motion.</thinking><answer>{"structural_stability": 3.0, "physical_plausibility": 4.5, "motion_coherence": 3.5}</answer>The code repository includes a GRPO fine-tuning script:
bash
python -m pip install -r checkpoints/MotionInsight-8B/requirements/train.txtATTN_IMPLEMENTATION=sdpa bash training/run_grpo.sh \ checkpoints/MotionInsight-8B /path/to/train.jsonl outputs/motioninsight-grpoSee the repository README for the input format and preprocessing setup.
Model weights use Apache-2.0. Preprocessing dependencies retain their own licenses; see NOTICE.
bibtex
@misc{zhan2026motioninsightdiagnosingobjectmotion, title = {MotionInsight: Diagnosing Object Motion Deficiencies in Generated Videos}, author = {Jiahao Zhan and Yongrui Ma and Qunliang Xing and Xuanyu Zhang and Jingqi Tong and Junlin Li and Li zhang and Shijie Zhao and Tianfan Xue}, year = {2026}, eprint = {2609.37030}, archivePrefix = {arXiv}, primaryClass = {cs.CV}, url = {https://arxiv.org/abs/2609.37030},}git clone --recurse-submodules https://github.com/JohnZhan2023/MotionInsight.gitcd MotionInsightpython3.10 -m venv .venv-modelsource .venv-model/bin/activatepython -m pip install --upgrade pippython -m pip install huggingface-hub==0.36.2hf download JohnZhan/MotionInsight-8B --local-dir checkpoints/MotionInsight-8Bpython -m pip install torch==2.11.0 torchvision==0.26.0 \ --index-url https://download.pytorch.org/whl/cu126python -m pip install -r checkpoints/MotionInsight-8B/requirements/inference.txtpython scripts/patch_transformers.pypython inference.py \ --model_name_or_path checkpoints/MotionInsight-8B \ --video_path /path/to/video.mp4 \ --object_motion_path /path/to/object_motion.pt \ --camera_motion_path /path/to/camera_motion.pt \ --target "tennis ball" \ --output_path outputs/prediction.jsonl<thinking>Diagnostic reasoning about the target object's motion.</thinking><answer>{"structural_stability": 3.0, "physical_plausibility": 4.5, "motion_coherence": 3.5}</answer>python -m pip install -r checkpoints/MotionInsight-8B/requirements/train.txtATTN_IMPLEMENTATION=sdpa bash training/run_grpo.sh \ checkpoints/MotionInsight-8B /path/to/train.jsonl outputs/motioninsight-grpo@misc{zhan2026motioninsightdiagnosingobjectmotion, title = {MotionInsight: Diagnosing Object Motion Deficiencies in Generated Videos}, author = {Jiahao Zhan and Yongrui Ma and Qunliang Xing and Xuanyu Zhang and Jingqi Tong and Junlin Li and Li zhang and Shijie Zhao and Tianfan Xue}, year = {2026}, eprint = {2609.37030}, archivePrefix = {arXiv}, primaryClass = {cs.CV}, url = {https://arxiv.org/abs/2609.37030},}