Skip to content

Issues: LLaVA-VL/LLaVA-NeXT

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

254 Open 74 Closed
254 Open 74 Closed
Author
Filter by author
Loading
Label
Filter by label
Loading
Use alt + click/return to exclude labels
or ⇧ + click/return for logical OR
Projects
Filter by project
Loading
Milestones
Filter by milestone
Loading
Assignee
Filter by who’s assigned
Sort
Sort by
Newest Oldest Most commented Least commented Recently updated Least recently updated Best match
Most reactions
👍 👎 😄 🎉 😕 ❤️ 🚀 👀

Issues list

finetune llava-onevision-qwen2-0.5b-si on custom dataset.
#368 opened Dec 18, 2024 by yanbai1993
Can the model input be text only?
#367 opened Dec 17, 2024 by ererdewubudesi
About pretrain script and data of LLaVA-NeXT
#366 opened Dec 17, 2024 by ligeng0197
When using LLaVA-Llama3 for batch inference with the generate function, the results are incorrect.
#365 opened Dec 15, 2024 by smile-struggler
catastrophic forgetting
#364 opened Dec 14, 2024 by sailfish009
About llava-video dataset
#363 opened Dec 13, 2024 by jun0wanan
4
Bugging Results? Repetitive Outputs and Long Inference Times with LLaVA-Video-7B-Qwen2
#362 opened Dec 10, 2024 by Rachel0901
Training LLaVA with the Liger kernel results in degraded performance.
#361 opened Dec 10, 2024 by y-rok
Issue with 4-bit Quantization for LLaVA-NeXT-Video-32B Model on A100-40GB GPU
#360 opened Dec 9, 2024 by Rachel0901
Loss=0 and grad_norm=nan when fine-tuning llava-v1.5-7b using dpo.sh
#359 opened Dec 5, 2024 by Na-nata
Question about Resolution Settings in LLAVA-OV Stage 1.5 Training
#358 opened Dec 4, 2024 by riariam
Where I can find data: "/data/llava_video/shareVideoGPTV/dpo/sft_dpo_17k.jsonl"
#357 opened Dec 2, 2024 by princepride
ValueError: Llama 3 tokenizer is not available. Make sure you have the necessary permissions.
#356 opened Nov 30, 2024 by sjghh
2
size mismatch for vision_model.post_layernorm.weight
#355 opened Nov 28, 2024 by zhanghang1995
8
Weird output for MSVD-QA dataset
#354 opened Nov 28, 2024 by oximi123
1
Missing mm_projector.bin in LLaVA-Video series
#353 opened Nov 28, 2024 by kpc0810
1
del tokenizer doesn't work for me in OOM
#352 opened Nov 28, 2024 by goodstudent9
Is the dataset size 1000(images) too small for both LLaVA-NeXT and OneVision?
#351 opened Nov 26, 2024 by Bleking
Warning: You are using a model of type llava to instantiate a model of type llava_qwen. This is not supported for all configurations of models and can yield errors.
#350 opened Nov 25, 2024 by kydxh
1
Where is the training script for mid_stage?
#348 opened Nov 22, 2024 by weiwei1206
Big fluctuations in training-loss
#347 opened Nov 22, 2024 by vir-95
LLaVA-Onevision: post_layernorm in SiglipVisionTransformer takes no effect in the forward
#346 opened Nov 22, 2024 by dyou-dev
Version of lmms-eval for llava-next image
#344 opened Nov 20, 2024 by EchoDreamer
After the mm_vision_tower and mm_mlp_adapter of llava-onevision-qwen2-7b-ov were fine-tuned, the model parameter shape did not match after the fine-tuning
#342 opened Nov 18, 2024 by zyf8818
4
Can the inference code of LLAVA-OV be modified to support batch inference? Single-instance inference is a bit slow.
#341 opened Nov 18, 2024 by cyinen
ProTip! Type g p on any issue or pull request to go back to the pull request listing page.