Video-Text-to-Text
Transformers
Safetensors
qwen3_vl
image-text-to-text
video
temporal-grounding
qwen3-vl
Instructions to use MCG-NJU/TimeLens2-4B with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- Transformers
How to use MCG-NJU/TimeLens2-4B with Transformers:
# pip install -U transformers accelerate # Load model directly from transformers import AutoProcessor, AutoModelForMultimodalLM processor = AutoProcessor.from_pretrained("MCG-NJU/TimeLens2-4B") model = AutoModelForMultimodalLM.from_pretrained("MCG-NJU/TimeLens2-4B", device_map="auto") - Notebooks
- Google Colab
- Kaggle
Commit History
30 days ago
-
Update TimeLens2 citation
Verified signaturef5dbceb
-
docs: add TimeLens2 paper citation
Verified signatureddbb6cb
-
Update README.md
Verified signaturebe1e11f
-
Highlight state-of-the-art performance
Verified signaturecb36f38 -
Use non-LFS benchmark SVG
Verified signaturea9a0621 -
Fix benchmark image rendering
Verified signaturecc8b050 -
Clarify TimeLens-Bench re-annotations
Verified signature8df5c62 -
Align TimeLens2 highlight with top divider
Verified signature490494a -
Remove Gemini 3 Pro and enlarge benchmark typography
Verified signature9e82ff6 -
Widen benchmark table for full-width display
Verified signaturee552d46 -
Remove outer card and enlarge benchmark table typography
Verified signaturecba555d -
Refine benchmark table layout and move TimeLens2 first
Verified signature87058f3 -
Replace benchmark bars with Claude-style results table
Verified signaturead678f6 -
Redesign benchmark chart and update grounding prompt
Verified signature382369b -
Add TimeLens2-4B model card
Verified signaturee11c6eb
-
Delete training_args.bin
Verified signature9060200 -
Delete trainer_state.json
Verified signature00983d1 -
Upload checkpoint-Timelens2-4B
Verified signature73bf266 -
initial commit
Verified signature67d6e5d