Repository Issue Activity (beta)

#513 Does llava-onevision-qwen2-7b-ov-chat support multiple video inputs

#508 LoRA Logic Bug Report:mm_tunable_parts + LoRA = Full-Model Training (Unintended)

#196 High Memory Usage in DataLoader Workers Leading to Out-of-Memory (OOM)

#494 non-meta parameter vs meta parameter

#504 Training crash on `main`: `NameError: name 'math' is not defined` in `llava_trainer.py`

#503 [Feature Request] Publish LLaVA-NeXT to PyPI for easier integration

#317 transformers-4.40.0.dev0

#454 Some questions about the test results of LLaVA-OneVision (multi-images input)and LLaVA-Video (video input)

#210 Use the llava-onevision weights as starting point for finetuning on custom dataset

#472 Error when using locally downloaded weights of google/siglip-so400m-patch14-384 from Hugging Face

#188 ValueError: Unknown vision tower: google/siglip-so400m-patch14-384

#501 video-to-text with vLLM

#500 Support stable release of `transformers` in `requirements.txt`

#499 Cannot reproduce the training of LLavaOnevision

#497 Can we run the LLaVA-v1.5-7B model on our environment?

#420 Get N/A when evaluating lmms-lab/llava-onevision-qwen2-7b-ov on mathvista_testmini, docvqa_test, and infovqa_test

#496 Trouble setting up lmms-eval

#486 Blackwell GPU support for PyTorch

#495 Request for LLaVA-OneVision Model & Data Specifications

#97 Any plans to support vLLM?

#249 dpo_ov7b.sh imports data_processing which is missing

#174 cauldron dataset

#483 Training fails with tons of missing imports in llava_trainer.py

#492 Setting different learning rate for different modules

#419 Multi-node job: resume_from_checkpoint incurs error: assert len(self.ckpt_list) > 0