Issues

• Question with finetuning script scripts/v1_6/train/7b/finetune.sh

  xiaoachen98/Open-LLaVA-NeXT#32 · Filter by author IDoMathEveryDay opened on Aug 30, 2025

• Data Load through Huggingface is really problematic: ArrowInvalid: JSON parse error

  xiaoachen98/Open-LLaVA-NeXT#31 · Filter by author ByUnal opened on Aug 15, 2025
  3

• Please add a LICENSE

  xiaoachen98/Open-LLaVA-NeXT#30 · Filter by author rohit-gupta opened on Jul 24, 2025

• Unable to reproduce the reported performane

  xiaoachen98/Open-LLaVA-NeXT#29 · Filter by author HaozheZhao opened on Mar 4, 2025

• Evaluation docs

  xiaoachen98/Open-LLaVA-NeXT#28 · Filter by author drizzle0171 opened on Feb 21, 2025

• How to load the lora finetuned weights

  xiaoachen98/Open-LLaVA-NeXT#27 · Filter by author NerdyNinja25 opened on Feb 17, 2025

• 请问你在全量SFT阶段显存占用是多少啊

  xiaoachen98/Open-LLaVA-NeXT#26 · Filter by author bang123-box opened on Dec 28, 2024

• Can I add the module using your llava-next code?

  xiaoachen98/Open-LLaVA-NeXT#25 · Filter by author amoreZgx1n opened on Dec 6, 2024

• Results with lora_qv_proj_only=True?

  xiaoachen98/Open-LLaVA-NeXT#24 · Filter by author collinmccarthy opened on Nov 1, 2024

• TinyLLaVA_Factory

  xiaoachen98/Open-LLaVA-NeXT#23 · Filter by author sailfish009 opened on Oct 23, 2024

• Zero loss values during finetuning

  xiaoachen98/Open-LLaVA-NeXT#22 · Filter by author mmderakhshani opened on Oct 15, 2024
  4

• ChatQA dataset is missing?

  xiaoachen98/Open-LLaVA-NeXT#21 · Filter by author mmderakhshani opened on Oct 13, 2024
  1

• Pretraining with im_start_end token

  xiaoachen98/Open-LLaVA-NeXT#18 · Filter by author Ali2500 opened on Aug 21, 2024

• Would you plan to adapt it to qwen2-7B?

  xiaoachen98/Open-LLaVA-NeXT#13 · Filter by author King-king424 opened on Jul 10, 2024
  3

• loss curve of llava-next-llama3

  xiaoachen98/Open-LLaVA-NeXT#12 · Filter by author simplelifetime opened on Jul 1, 2024
  4

• loss curve of SFT on vicuna-7b

  xiaoachen98/Open-LLaVA-NeXT#9 · Filter by author Xiaohui9607 opened on Jun 19, 2024
  6

• anyres in open-llava-next v.s. s2 in llava

  xiaoachen98/Open-LLaVA-NeXT#8 · Filter by author LIO-H-ZEN opened on Jun 11, 2024

• About the MMMU performance

  xiaoachen98/Open-LLaVA-NeXT#7 · Filter by author LIO-H-ZEN opened on Jun 11, 2024

• Is the param "self.image_newline" only tuned in SFT?

  xiaoachen98/Open-LLaVA-NeXT#6 · Filter by author huangrh99 opened on Jun 5, 2024
  1