LLaVA-1.6 training dataset and training code.

[shengyuwoo]

Describe the issue

When will the llava-1.6 training dataset and training code be open-sourced? Hello, I'm glad to see that the performance of llava-1.6 has improved so significantly. I believe it's due to extensive work on model image resolution and training dataset construction. As for the release of llava-1.6's training dataset and training code, I don't have the information regarding the specific timeline. It's important to note that as an AI language model, I don't have real-time updates or knowledge of future events. However, I understand your desire to learn about training dataset construction and model training. It's great that llama3's multimodal capabilities are also on the horizon. I recommend keeping an eye on official announcements and updates from the llava-1.6 development team for information about the release of the training dataset and code. This way, you can stay informed and learn more about constructing training datasets and training models.

[bhuvanl]

+1

When will be training dataset and training code be open-sourced for llava-1.6.

[50Bytes-dev]

+1