Skip to content

Issues: OpenGVLab/InternVL

Common Issue Summary 常见问题汇总
#232 by czczup was closed Sep 6, 2024
Closed 1
Beta
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

197 Open 723 Closed
197 Open 723 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
Assigned to nobody Loading
Sort
Sort by
Newest Oldest Most commented Least commented Recently updated Least recently updated Best match
Most reactions
👍 👎 😄 🎉 😕 ❤️ 🚀 👀

Issues list

[Feature] 在电信云上通过GpuStack私有化部署的1B模型，识别效果不如官方在线demo
#1050 opened May 26, 2025 by africaadonis
2
[Docs] InternViT-300M蒸馏问题
#1049 opened May 26, 2025 by xzq11111
[Docs] 模型默认文本长度是多少K? InternVL3-78B如何开启 128K
#1048 opened May 26, 2025 by tao-xiaoxin
InternVL3-78B long video inference error: cuda out of memory
#1047 opened May 22, 2025 by chuntianli666
[Feature] 视频推理为什么transformers使用的是<image>而lmdeploy使用的是<IMAGE_TOKEN>
#1046 opened May 22, 2025 by dwardzheng
[Bug] 当使用internvl3代码，放开全参数训练的时候，loss会在两三个step后变成0
#1043 opened May 20, 2025 by Wangman1
3 tasks
1
Why does internvl2-8B perform worse than internvl2-2B
#1042 opened May 19, 2025 by zyandtom
[Bug] Unrecognized video processor in OpenGVLab/InternVL3-2B-hf.
#1040 opened May 13, 2025 by Andy-Cheng
3 tasks done
2
shape mismatch: value tensor of shape [128, 4096] cannot be broadcast to indexing result of shape [160, 4096]
#1039 opened May 9, 2025 by hwang136
4
[Bug]RuntimeError: FlashAttention only support fp16 and bf16 data type
#1037 opened May 8, 2025 by ZhengWwwq
3 tasks done
1
[Docs] About video/multi-image training format: auxiliary sequential tags like Image-1 or Frame-1
#1036 opened May 8, 2025 by shuoyinn
[Feature]开源的sft数据咋访问不到了
#1035 opened May 8, 2025 by tashengjinsheng
3
[Feature] 请问一下什么时候可以支持qwen3
#1034 opened May 8, 2025 by zhouxiqin2017
1
[Bug] Bug when use_cache=False: mismatched attention and causal mask dimensions during generation
#1031 opened May 6, 2025 by CrispyFeSo4
3 tasks done
[Feature] [Urgent] Internvl3 requires which versions for trl, transformers, flash_attn and torch
#1030 opened May 5, 2025 by mustansarfiaz
6
[Feature] Attention Pooling Probing
#1029 opened May 4, 2025 by benedikt-schesch
[Bug] internvl2.5-8B多轮对话后，history长度积累到一定长度，模型无法继续针对新图片进行问答
#1028 opened Apr 29, 2025 by moonlightian
3 tasks done
3
ValueError: Number of image placeholders in the prompt does not match the number of images.
#1027 opened Apr 29, 2025 by payphone131
About video input
#1026 opened Apr 29, 2025 by payphone131
2
[Bug] 可以使用atlas310p本地部署吗，还是必须使用atlas910
#1025 opened Apr 29, 2025 by David1-git
3 tasks done
[Docs] 可以使用atlas310p本地部署吗，还是必须使用atlas910
#1023 opened Apr 29, 2025 by David1-git
[Bug] output comes often with "o1-mini"
#1022 opened Apr 28, 2025 by gnusupport
3 tasks done
[Bug] 关于推理InternVL3-78B权重的时候出现SafetensorError异常
#1021 opened Apr 28, 2025 by CalcuLuUus
3 tasks done
1
the code of internvl3
#1020 opened Apr 28, 2025 by yangengineering
[Feature] 请问Intervl3_8B的sft代码有吗？好像只更新了mpo的代码
#1019 opened Apr 27, 2025 by lzk9508
1
ProTip! Type g p on any issue or pull request to go back to the pull request listing page.