Akshata

issue while finetuning DocOwl1.5-Omni on dataset

4

The training does not start..my memory is completely occupied but GPU is at 0%. Screenshot attached below. Pls help . 

requirements file

1

would be great if the team could point me to the requirements file for finetuning .

IndexError: piece id is out of range

1

Hi getting this error while running inference: Reloading the model helps, but it is extremely time consuming. `The attention mask and the pad token id were not set. As a...

[Feature]: LoRA support for InternVLChatModel

10

### Your current environment vllm version = 0.6.1 ### Model Input Dumps _No response_ ### 🐛 Describe the bug The output of `command:` vllm version = 0.6.1. InternVLChat is in...

Question - What is a good prompt to get the output of the image in a structured format.

2

Since qwen2.5 VL is capable of maintaining layout information; I was wondering what is a good prompt to get output in a structured format. For example, given image of an...

[Enhancement] Exploring modular, DSPy-style prompting to support scale

1

Hi team, I wanted to suggest a possible enhancement to prompting by structuring prompts in a more modular, DSPy-style format. Instead of encoding all logic in raw natural language, we...

This issue aims to refactor the current Dockerfile to use a multistage build approach. The motivation for this change is to: 1. Reduce final image size by excluding unnecessary build-time...