Video-Text-to-Text
Transformers
English
video
video-question-answering
multimodal
vision-language
qwen3-vl
inference-time
frame-selection
clip
Instructions to use commandeaw/DW-KhotTaeVL-2B-QueryFrames with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- Transformers
How to use commandeaw/DW-KhotTaeVL-2B-QueryFrames with Transformers:
# pip install -U transformers accelerate # Load model directly from transformers import AutoModel model = AutoModel.from_pretrained("commandeaw/DW-KhotTaeVL-2B-QueryFrames", device_map="auto") - Notebooks
- Google Colab
- Kaggle
Commit History
-
Initial release: DW-KhotTaeVL-2B-QueryFrames v1.0
Verified signatureQuery-aware frame selection wrapper for Qwen3-VL-2B-Instruct. Wild mode: 64.3% on Video-MME mini 300Q (+7.3pp vs stock 57.0%). Benchmark mode: 66.3% (+9.3pp), 12/12 task buckets non-negative. Zero trainable parameters, no model weights modified. Built on Qwen/Qwen3-VL-2B-Instruct (Apache 2.0). Frame scorer: openai/clip-vit-large-patch14 (MIT). Author: Deaw (HF: @commandeaw).84c8a9d