StartHua/Comfyui_CXH_joy_caption

Issues

• # ComfyUI Error Report ## Error Details - **Node Type:** Joy_caption - **Exception Type:** OSError - **Exception Message:** Error no file named pytorch_model.bin, model.safetensors, tf_model.h5, model.ckpt.index or flax_model.msgpack found in directory D:/comfyui/models/clip\siglip-so400m-patch14-384.

  #99 opened 2 months ago by wbqb986
  1

• florence_nodes.py 找不到本地LLM文件夹

  #112 opened 10 days ago by Nutingnon
  0

• Joy_caption_alpha_run You can't move a model that has some modules offloaded to cpu or disk.

  #90 opened 3 months ago by lestersssss
  7

• `.to` is not supported for `4-bit` or `8-bit` bitsandbytes models

  #111 opened 15 days ago by quanquan-dane
  0

• mac版本的comfyui客户端，运行了requirement.txt后，还是无法运行节点

  #110 opened 16 days ago by lcw552003428
  0

• Error no file named pytorch_model.bin, model.safetensors, tf_model.h5, model.ckpt.index or flax_model.msgpack found in directory D:/comfyui/models/clip\siglip-so400m-patch14-384

  #101 opened 2 months ago by 158zd
  2

• Download and load the Florence2 model

  #109 opened 18 days ago by DNPMBHC
  0

• CHX_Florence2Run节点需要输入一个seed，如何连接？之前不用的啊。

  #108 opened 23 days ago by kame-boop
  0

• This modeling file requires the following packages that were not found in your environment: flash_attn. Run `pip install flash_attn`

  #91 opened 3 months ago by mr-bob-chang
  5

• 批量打标使用unsloth/Meta-Llama-3.1-8B-bnb-4bit，txt文件会出现大量重复的内容

  #107 opened 25 days ago by 2259819800
  0

• Joy_caption_alpha_load Descriptors cannot be created directly. If this call came from a _pb2.py file, your generated code is out of date and must be regenerated with protoc >= 3.19.0. If you cannot immediately regenerate your protos, some other possible workarounds are: 1. Downgrade the protobuf package to 3.20.x or lower. 2. Set PROTOCOL_BUFFERS_PYTHON_IMPLEMENTATION=python (but this will use pure-Python parsing and will be much slower). More information: https://developers.google.com/protocol-buffers/docs/news/2022-05-06#python-updates

  #85 opened 3 months ago by 108863258
  1

• GGUF ?

  #106 opened a month ago by Metal-dude
  0

• 提示加载 sentencepiece 相关文件时出现了重复的 proto 文件。求教如何解决

  #77 opened 3 months ago by qingdengke88
  4

• Is it possible to run it on GPU0 and comfy on GPU1?

  #105 opened a month ago by nikolaiusa
  0

• Joy_caption Error(s) in loading state_dict for ImageAdapter: Unexpected key(s) in state_dict: "other_tokens.weight".

  #72 opened 3 months ago by xiewci
  6

• batch image interrogate error

  #71 opened 3 months ago by hopeyan476868
  1

• Expected all tensors to be on the same device, but found at least two devices, cpu and cuda:0! (when checking argument for argument tensors in method wrapper_CUDA_cat)

  #104 opened a month ago by 1247862674
  0

• JoyCaption [Errno 30] Read-only file system: '/comfyui'

  #103 opened a month ago by hoonkies
  0

• 大佬，啥时候加 ic-lora 格式的提示词啊？

  #102 opened 2 months ago by 807502278
  0

• 依赖版本太老，可以自己修改吗？

  #70 opened 3 months ago by K-O-N-B
  2

• Joy_caption节点报错 expected str, bytes or os.PathLike object, not NoneType

  #92 opened 3 months ago by mr-bob-chang
  1

• required input is missing: seed

  #100 opened 2 months ago by realyn
  2

• 错误：在目录 C:/comfyui/models/clip\siglip-so400m-patch14-384 中找不到名为 pytorch_model.bin、model.safetensors、tf_model.h5、model.ckpt.index 或 flax_model.msgpack 的文件。

  #60 opened 4 months ago by jiej32228
  3

• CXH_Florence2Run节点在linux里有问题

  #80 opened 3 months ago by 807502278
  1

• 大佬，请问如何实现Joy_caption_v2批量打标签呢？

  #98 opened 2 months ago by YoucanBaby
  1

• transformers版本问题!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!

  #73 opened 3 months ago by poyingshihuang
  2

• 使用Meta-Llama-3.1-8B-bnb-4bit加载Joycaption的时候`rope_scaling` must be a dictionary with two fields, `type` and `factor`, got {'factor': 8.0, 'high_freq_factor': 4.0, 'low_freq_factor': 1.0, 'original_max_position_embeddings': 8192, 'rope_type': 'llama3'}这个报错

  #67 opened 3 months ago by Pondowner857
  2

• 有个小建议

  #97 opened 2 months ago by 807502278
  0

• control character (\u0000-\u001F) found while parsing a string at line 16971

  #96 opened 2 months ago by xuhuanquan
  0

• V2不能使用

  #95 opened 2 months ago by chenpipi0807
  0

• NEED SDWebUI version！！！

  #94 opened 3 months ago by Zhuangvictor0
  0

• When multiple GPUs report errors.

  #93 opened 3 months ago by tiandaoyuxi
  0

• Joy_caption_alpha_run 加载错误

  #89 opened 3 months ago by LiuGe126
  2

• 除零报错Windows fatal exception: int divide by zero

  #88 opened 3 months ago by sakusankun
  0

• Remove specific requirement versions

  #87 opened 3 months ago by EnragedAntelope
  0

• Joy_caption_alpha_prompt这个节点从name开始所有选项错了一位。

  #83 opened 3 months ago by DragonQuix
  1

• 安装依赖的时候，其中llama-cpp-python==0.2.89这个安装错误啊

  #86 opened 3 months ago by doorle
  1

• 关于chat

  #84 opened 3 months ago by Chengym2023
  0

• ValueError: list.remove(x): x not in list -- Error when using Florence models

  #79 opened 3 months ago by patriciagomesoo
  3

• joy caption alpha two support

  #75 opened 3 months ago by soldivelot
  2

• Comfyui_image2prompt node, delete this to solve the problem

  #81 opened 3 months ago by poyingshihuang
  0

• 提示protobuf版本问题，但是检测系统版本又是对的，不知道问题在哪？求教

  #76 opened 3 months ago by qingdengke88
  1

• support llama3.2 please

  #78 opened 3 months ago by Lagrebanana
  0

• Some modules are dispatched on the CPU or the disk. Make sure you have enough GPU RAM to fit the quantized model.

  #74 opened 3 months ago by yatoubusha
  1

• joy caption错误，用的bnb-4bit模型

  #68 opened 3 months ago by K-O-N-B
  6

• How to fix this ?

  #69 opened 3 months ago by vihangasa14
  0

• 能不能加一个检测到已经拥有tag文件是否可以选择覆盖或者追加按钮，因为这真的很需要

  #65 opened 3 months ago by lixida123
  1

• vision_config is None, using default vision config Unused kwargs: ['_load_in_4bit', '_load_in_8bit', 'quant_method']. These kwargs are not used in . `low_cpu_mem_usage` was None, now set to True since model is quantized.

  #63 opened 4 months ago by water110
  0

• What's the input parameter temperature?

  #62 opened 4 months ago by ykhasia
  0

• This error originates from a subprocess, and is likely not a problem with pip.

  #61 opened 4 months ago by kuzhushidai
  0