deepseek-ai/DeepSeek-VL

Issues

• Does not start

  #29 opened 10 months ago by Troceleng
  1

• 多图的数量上限

  #54 opened 7 months ago by JUNJIE99
  2

• 什么时候可以上开放平台

  #63 opened 23 days ago by 466579141
  0

• ImportError: cannot import name 'load_pil_images_from_urls' from 'deepseek_vl.utils.io' (/hpc2hdd/home/jsu360/DeepSeek-VL/deepseek_vl/utils/io.py)

  #62 opened a month ago by jsu360
  0

• ValueError: The checkpoint you are trying to load has model type `multi_modality` but Transformers does not recognize this architecture. This could be because of an issue with the checkpoint, or because your version of Transformers is out of date.

  #60 opened 3 months ago by Y-PanC
  2

• default_conversation class is not implemented.

  #61 opened 2 months ago by HaithemH
  0

• training cost of deepseek-vl-1.3B

  #59 opened 5 months ago by shufangxun
  0

• TypeError: Object of type AlignerConfig is not JSON serializable

  #58 opened 6 months ago by JingyuanHan
  1

• dataset format of pretraining stage

  #56 opened 6 months ago by annopackage
  0

• error when i run inference.py

  #37 opened 9 months ago by Yangfan-96
  6

• How to get feature representation vector of input image?

  #55 opened 7 months ago by lukywangwang
  0

• How can I get a GenerateOutput after language_model.generate() ?

  #53 opened 8 months ago by ApolloRay
  0

• RuntimeError: "triu_tril_cuda_template" not implemented for 'BFloat16'

  #52 opened 8 months ago by sulaiman1988
  1

• If I want to run and use in ollama, I can import the GGUF version into ollama if it is available

  #49 opened 9 months ago by alex13by
  1

• Batching

  #48 opened 9 months ago by SinanAkkoyun
  7

• Finetuning

  #35 opened 10 months ago by univa-JASON
  2

• 支持ollama部署吗

  #50 opened 9 months ago by yuanjie-ai
  0

• 是否又评估过OCR能力？

  #8 opened 10 months ago by lucasjinreal
  5

• An error is reported for multi-GPU inference

  #46 opened 9 months ago by Athicbliss
  0

• LMDeploy support accelerating DeepSeek VL models now!!! :rocket:

  #39 opened 9 months ago by AllentDan
  2

• grdio加载模型卡住

  #45 opened 9 months ago by LianghuiGuo
  0

• What's the difference of base and chat model?

  #43 opened 9 months ago by BlueBlueFF
  1

• question: help needed for arch clarification

  #44 opened 9 months ago by SinanAkkoyun
  0

• 4-bit quant?

  #28 opened 10 months ago by nmandic78
  5

• What tasks in training for image-text interleaved data?

  #41 opened 9 months ago by BlueBlueFF
  1

• tips for running the model in FP16 on 24GB GPU

  #42 opened 9 months ago by adamo1139
  0

• 模型重复输出最后一句话，是什么原因呢，有办法能解决吗

  #40 opened 9 months ago by Yangfan-96
  1

• How to convert it to GGUF/GGML for general use?

  #38 opened 9 months ago by YuanfengZhang
  0

• pretrain 模型通过什么prompt 能激发OCR能力，能否输出坐标信息

  #36 opened 9 months ago by paulpaul91
  0

• Multi-image mixed input

  #20 opened 10 months ago by TechxGenus
  4

• LLaVA1.6-GPT4V是什么数据集呢？

  #33 opened 10 months ago by mzfive
  1

• Adding support for quantization and fast inference server

  #34 opened 10 months ago by rostandk
  0

• 'FieldInfo' object has no attribute 'required'

  #15 opened 10 months ago by White-Friday
  1

• Support for M1 Mac, or non-cuda devices

  #11 opened 10 months ago by mattkanwisher
  3

• Can you address the work you have referenced?

  #19 opened 10 months ago by lucasjinreal
  5

• ImportError: cannot import name 'Mapping' from 'collections'

  #4 opened 10 months ago by NickLucche
  3

• deepseek-vl系列的微调支持 (finetune)

  #10 opened 10 months ago by Jintao-Huang
  6

• RuntimeError: cuDNN error: CUDNN_STATUS_NOT_INITIALIZED

  #30 opened 10 months ago by linny2002
  0

• Fine-tuning Script

  #6 opened 10 months ago by TechxGenus
  7

• How to finetune and serve the finetuned model using VLLM like tools

  #2 opened 10 months ago by monksgoyal
  3

• Gradio issue on colab notebook T4 gpu

  #25 opened 10 months ago by clarencerat
  0

• ERROR: Package 'deepseek-vl' requires a different Python: 3.11.8 not in '<3.10,>=3.8'

  #24 opened 10 months ago by kakaxixx
  1

• 是否评估过有作为web或者windows的agent的潜力？

  #23 opened 10 months ago by DirtyKnightForVi
  0

• About the preprocessor way in high and low input?

  #17 opened 10 months ago by lucasjinreal
  0

• 请问是否后期会实现目标检测功能

  #18 opened 10 months ago by whysirier
  0

• ROCm Support

  #14 opened 10 months ago by hchasens
  1

• 请问什么时候可以开源微调代码

  #9 opened 10 months ago by cdxzyc
  1

• text-generation-webui是否支持

  #7 opened 10 months ago by newstargo
  0

• Does the sam-branch use the Vary initialization for dense OCR？

  #3 opened 10 months ago by Ucas-HaoranWei
  0