1 Issue closed by 1 person

• qwen2-vl-72B in V100 gptq error CUDA out of memory

  #619 closed Dec 26, 2024

5 Issues opened by 5 people

• 请教下vision-rope和m-rope该如何理解

  #621 opened Dec 26, 2024

• What are the algorithms behind the reasoning ability of QVQ and QWQ?

  #620 opened Dec 26, 2024

• Is there a plan to open source the Naive Dynamic Resolution ViT base model?

  #618 opened Dec 26, 2024

• Hidden states of image patches

  #617 opened Dec 26, 2024

• Increase batch size results in a non-linear increase in computation time

  #616 opened Dec 22, 2024

17 Unresolved conversations

Sometimes conversations happen on old items that aren’t yet closed. Here is a list of all the Issues and Pull Requests with unresolved conversations.

• 论文里的例子无法复现

  #605 commented on Dec 23, 2024 • 0 new comments

• Qwen2vl vllm grounding任务效果不如transformers推理

  #601 commented on Dec 24, 2024 • 0 new comments

• 关于 Qwen2-VL-7B-Instruct-GPTQ-Int4部署应用的问题

  #139 commented on Dec 24, 2024 • 0 new comments

• 使用量化模型Qwen2-VL-7B-Instruct-AWQ 安装autoawq以后 出现cannot import name 'shard_checkpoint' from 'transformers.modeling_utils' (F:\ai\qwen_vl_simple_api-main\py310\lib\site-packages\transformers\modeling_utils.py)错误

  #589 commented on Dec 25, 2024 • 0 new comments

• qwen2Vl vllm框架如何添加视频内容

  #540 commented on Dec 25, 2024 • 0 new comments

• qwen2Vl vllm框架如何添加视频内容

  #539 commented on Dec 25, 2024 • 0 new comments

• Different Commands for SFT

  #290 commented on Dec 25, 2024 • 0 new comments

• 怎么让其做目标检测，输出图像中物体的坐标？

  #588 commented on Dec 26, 2024 • 0 new comments

• IndexError: index 0 is out of bounds for dimension 0 with size 0 for " if cache_position is None or (cache_position is not None and cache_position[0] == 0):"

  #596 commented on Dec 26, 2024 • 0 new comments

• 关于使用AutoGPTQ量化的问题

  #863 commented on Dec 26, 2024 • 0 new comments

• visual grounding poor performance after fine-tuning

  #584 commented on Dec 26, 2024 • 0 new comments

• How to use bbox in SFT/微调中的定位框问题

  #105 commented on Dec 26, 2024 • 0 new comments

• 请问支持视觉定位吗？

  #9 commented on Dec 26, 2024 • 0 new comments

• 视觉定位refcoco评测和官方结果稍有误差，请问官方的prompt是什么

  #536 commented on Dec 26, 2024 • 0 new comments

• 训练的时候，batch 内的所有图片会拼到一起？

  #595 commented on Dec 27, 2024 • 0 new comments

• 通过VLLM推理，可以支持多图理解吗？

  #612 commented on Dec 27, 2024 • 0 new comments

• 添加Flash-attention2后，使用Qwen2-7B-VL进行推理，会报错

  #565 commented on Dec 29, 2024 • 0 new comments