Issues

• run peft_infer.py and get TypeError: TextEncodeInput must be Union[TextInputSequence, Tuple[InputSequence, InputSequence]]

  zai-org/CogVLM2#219 · Filter by author flyingcornjuice opened on Aug 21, 2025

• AttributeError: 'CogVLMForCausalLM' object has no attribute '_extract_past_from_model_output'

  zai-org/CogVLM2#218 · Filter by author XuJianzhi opened on Jun 18, 2025
  1

• 如何对长视频进行理解

  zai-org/CogVLM2#217 · Filter by author SMK1311 opened on Jun 18, 2025

• 线上测试demo不可用，

  zai-org/CogVLM2#216 · Filter by author Crick-Zhou opened on Apr 11, 2025

• 文档写这么差，真的不考虑优化一下吗

  zai-org/CogVLM2#215 · Filter by author hecongyuan opened on Mar 13, 2025
  2

• 1-Click installers for Windows, RunPod, Massed Compute and ultra advanced CogVLM 2 Batch Processing Gradio APP with 4-bit quantization

  zai-org/CogVLM2#212 · Filter by author FurkanGozukara opened on Jan 17, 2025
  1

• 指定多张GPU运行报AttributeError: 'Params4bit' object has no attribute '_parameters'

  zai-org/CogVLM2#211 · Filter by author sf9ehf9fe opened on Jan 10, 2025

• 如何指定多张GPU卡跑，默认用一张，显示GPU内存不够

  zai-org/CogVLM2#208 · Filter by author sf9ehf9fe opened on Dec 26, 2024

• ImportError: /home/root/anaconda/envs/cogvlm2/lib/python3.10/site-packages/torch/lib/../../nvidia/cusparse/lib/libcusparse.so.12: symbol __nvJitLinkComplete_12_4, version libnvJitLink.so.12 not defined in file libnvJitLink.so.12 with link time reference

  zai-org/CogVLM2#207 · Filter by author sf9ehf9fe opened on Dec 26, 2024
  1

• TypeError: _extract_past_from_model_output() got an unexpected keyword argument 'standardize_cache_format'

  zai-org/CogVLM2#204 · Filter by author sjghh opened on Dec 16, 2024
  2

• 请问如何拿几张图片做fewshot

  zai-org/CogVLM2#203 · Filter by author Nand-Lu opened on Dec 2, 2024

• 用xinference在jetson agx orin上部署cogvlm2模型后，模型可以加载但是无法进行对话，报[address=0.0.0.0:38925, pid=1624735] object of type 'NoneType' has no len()

  zai-org/CogVLM2#202 · Filter by author GengyuXu opened on Nov 12, 2024

• Fine tune for detection

  zai-org/CogVLM2#201 · Filter by author ck-amrahd opened on Nov 12, 2024

• 论文中提到的对齐数据会开源吗

  zai-org/CogVLM2#197 · Filter by author ZhangLin009 opened on Oct 17, 2024

• Question about Caption Model

  zai-org/CogVLM2#195 · Filter by author zhiyuanyou opened on Oct 15, 2024
  1

• Question about expert_mask

  zai-org/CogVLM2#194 · Filter by author xiaojunjie opened on Oct 14, 2024
  1

• finetune cogvlm2-video

  zai-org/CogVLM2#192 · Filter by author yepzhang opened on Sep 29, 2024
  3

• Question about face-specific action analysis in video input

  zai-org/CogVLM2#189 · Filter by author YoungjaeDev opened on Sep 10, 2024

• CogVLM grounding training data

  zai-org/CogVLM2#187 · Filter by author Michael4933 opened on Sep 2, 2024

• Is there a plan to release the TQA dataset used to train CogVLM2-Video?

  zai-org/CogVLM2#186 · Filter by author peiliu0408 opened on Sep 2, 2024
  3

• 输入的图片尺寸过大，微调后的结果会有什么影响？

  zai-org/CogVLM2#185 · Filter by author WuChannn opened on Aug 29, 2024

• TypeError: GenerationMixin._extract_past_from_model_output() got an unexpected keyword argument 'standardize_cache_format with transformers==4.44.0

  zai-org/CogVLM2#181 · Filter by author 7801943 opened on Aug 16, 2024
  3

  

• 是否支持全参微调？

  zai-org/CogVLM2#180 · Filter by author WuChannn opened on Aug 15, 2024
  2

  

• 两张A100（共80G显存）测试openai_api_demo.py时报错OOM

  zai-org/CogVLM2#179 · Filter by author thu-yn opened on Aug 7, 2024
  2

  

• 当我在用A100运行微调代码的时候，出现torch.distributed.DistBackendError: NCCL error in: ../torch/csrc/distributed/c10d/ProcessGroupNCCL.cpp:1691, unhandled system error (run with NCCL_DEBUG=INFO for details), NCCL version 2.19.3 ncclSystemError: System call (e.g. socket, malloc) or external library call failed or device error.

  zai-org/CogVLM2#178 · Filter by author mindhunter01 opened on Aug 1, 2024