Search results

Clarification needed on the exact number of training epochs

Status: Open.

#11 In qizhou000/VisEdit;

· FryUoh opened on Mar 12, 2026

Hardware configuration used for running vead_train.py

Status: Open.

#10 In qizhou000/VisEdit;

· lxh633-crypto opened on Jan 14, 2026

您好，可以提供下下载llava-v1.5-7b的地址吗，总是会出现ValueError: The input provided to the model are wrong. The number of image tokens is 576 while the number of image given to the model is 1. This prevents correct indexing and breaks batch generation.错误

Status: Open.

#9 In qizhou000/VisEdit;

· zzidlezz opened on Nov 16, 2025

attention_mask is None type

Status: Open.

#8 In qizhou000/VisEdit;

· Wang-Z-W opened on Nov 4, 2025

Codebase for other baseline methods

Status: Open.

#6 In qizhou000/VisEdit;

· kailinjiang opened on Apr 8, 2025

How to deal with a process holding GPU memory after termination？

Status: Open.

#5 In qizhou000/VisEdit;

· zzFestinaLente opened on Apr 2, 2025

BLIP2-OPT-2.7B checkpoint

Status: Open.

#4 In qizhou000/VisEdit;

· BinjieW opened on Mar 21, 2025

SOTA Model Checkpoints

Status: Open.

#2 In qizhou000/VisEdit;

· TenChickens666 opened on Feb 23, 2025

Question for "llava-hf/llava-1.5-7b-hf"

Status: Open.

#1 In qizhou000/VisEdit;

· sev777 opened on Jan 6, 2025