Issues

• Verify evals on Papers with Code

  DAMO-NLP-SG/VideoLLaMA2#181 · opened on Aug 28, 2026

• When attn_implementation = 'eager', some of the logits values become NaN.

  DAMO-NLP-SG/VideoLLaMA2#178 · opened on Sep 29, 2025
  1

• Size mismatch error when training AV model with finetune script

  DAMO-NLP-SG/VideoLLaMA2#177 · opened on Sep 28, 2025
  3

• Audio–Video inference returns only "!!!!!!!"

  DAMO-NLP-SG/VideoLLaMA2#176 · opened on Sep 16, 2025
  2

• Failed to use base madel to inference.

  DAMO-NLP-SG/VideoLLaMA2#175 · opened on Sep 14, 2025

• Using the fine-tuned model for video inference, the output is null

  DAMO-NLP-SG/VideoLLaMA2#174 · opened on Jul 31, 2025
  1

• Installation of the package breaks for Python 3.12

  DAMO-NLP-SG/VideoLLaMA2#172 · opened on May 27, 2025

• Can not reproduce results on MVbench.

  DAMO-NLP-SG/VideoLLaMA2#171 · opened on Apr 9, 2025

• AVQA Result Different from The Reported Result.

  DAMO-NLP-SG/VideoLLaMA2#170 · opened on Mar 26, 2025
  1

• VideoLLaMA on CPU Server(without GPU or CUDA Support)

  DAMO-NLP-SG/VideoLLaMA2#169 · opened on Mar 25, 2025

• AVSD dataset

  DAMO-NLP-SG/VideoLLaMA2#168 · opened on Mar 24, 2025

• The inference results of the local deployment of lixin4ever/VideoLLaMA2-AV are inconsistent with the web DEMO results

  DAMO-NLP-SG/VideoLLaMA2#167 · opened on Mar 24, 2025

• Where is the LOSS function defined？

  DAMO-NLP-SG/VideoLLaMA2#166 · opened on Mar 18, 2025
  1

• Training Loss Stuck at Zero in VideoLLaMA2.1-7B-AV

  DAMO-NLP-SG/VideoLLaMA2#165 · opened on Mar 17, 2025
  1

• Demo not run

  DAMO-NLP-SG/VideoLLaMA2#164 · opened on Mar 16, 2025

• RuntimeError: CUDA error: an illegal memory access was encountered

  DAMO-NLP-SG/VideoLLaMA2#163 · opened on Mar 11, 2025
  1

• ValueError: CLIPVisionModel does not support Flash Attention 2.0 yet.

  DAMO-NLP-SG/VideoLLaMA2#162 · opened on Mar 5, 2025
  1

• How to train the model on 4 GPUs?

  DAMO-NLP-SG/VideoLLaMA2#160 · opened on Mar 1, 2025
  1

• why does transformer version matter so much ?

  DAMO-NLP-SG/VideoLLaMA2#157 · opened on Feb 21, 2025

• what is the difference between "VideoLLaMA2.1-7B-16F-Base" and "VideoLLaMA2.1-7B-16F" ?

  DAMO-NLP-SG/VideoLLaMA2#156 · opened on Feb 20, 2025

• How to Extract Video and Audio Features Using VideoLLaMA2?

  DAMO-NLP-SG/VideoLLaMA2#155 · opened on Feb 19, 2025
  1

• How to process multiple consecutive frames for fine-tuning VideoLLaMA2?

  DAMO-NLP-SG/VideoLLaMA2#154 · opened on Feb 19, 2025
  1

• Cannot load model in 4bit with VideoLLaMA2.1-7B-AV model

  DAMO-NLP-SG/VideoLLaMA2#151 · opened on Jan 28, 2025

• RuntimeError: Error reading datasets/videollava_pt/valley/023651_023700/1014646391.mp4

  DAMO-NLP-SG/VideoLLaMA2#150 · opened on Jan 25, 2025

• I want to batch process video files

  DAMO-NLP-SG/VideoLLaMA2#148 · opened on Jan 14, 2025