| Multilingual powerhouse — testing for mobile deployment | 3morixd | open | 0 | disc |
| "You are using a model of type deepseek_vl_v2 to instantiate a model of type DeepseekOCR2" Warning | AndreyPanfilov | open | 0 | disc |
| 什么时候适配50系显卡啊，sm_120，vllm使用也有问题，啥时候支持较新的版本 | xiaotu6e | open | 0 | disc |
| Is that deepseek-ocr-2 compatible for H200 | balajivasudevan | open | 0 | disc |
| Add ParseBench evaluation results | boyang-runllama | open | 0 | disc |
| Add ParseBench evaluation results | boyang-runllama | closed | 0 | disc |
| Rename README.md to ebook sur la monétisation de contenu sur Tik tok | Marquissalomon | open | 0 | disc |
| Rename README.md to ebook sur la monétisation de contenu sur Tik tok | Marquissalomon | open | 0 | disc |
| Duestions about the Stage 1 Encoder Pretraining | nero2023 | open | 0 | disc |
| config.kv_lora_rank is null? | SuHu753 | open | 0 | disc |
| 元宵节快乐，v4也该来了吧？ | luckjone | open | 0 | disc |
| How many Language? | aoiandroid | open | 0 | disc |
| V4? | 16dvnk | open | 0 | disc |
| Update conversation.py | HyzeAI | open | 0 | disc |
| Add OlmOCRBench evaluation results | nielsr | open | 0 | disc |
| Upload 5 files | Fuckubae | closed | 0 | disc |
| The model performs poorly. | wyb5210 | open | 0 | disc |
| Update README.md | Lawziet | open | 0 | disc |
| Update README.md to include the correct Arxiv link instead of linking to the github pdf | shokrydev | open | 0 | disc |
| Request a compatible version with newer transformer versions | AliKhajegiliM | open | 0 | disc |
| Link model to paper and update citation | nielsr | open | 0 | disc |
| What do you mean by "novel"? Google has been doing this for ages | davidboring | open | 0 | disc |
| Analysis on Prompt + Document Type Compatibility | ysong21 | open | 0 | disc |
| Installation Video and Testing - Step by Step | fahdmirzac | open | 0 | disc |
| Testing video and colab notebook, Issues with accuracy | ritheshSree | open | 0 | disc |
| ms-swift has supported inference, deployment, and fine-tuning of the DeepSeek-OCR-2 model. | hu5enpai | open | 0 | disc |
| Guide to Run + Fine-tune DeepSeek-OCR 2 | danielhanchen | open | 0 | disc |
| Model architectures ['DeepseekOCR2ForCausalLM'] are not supported with latest vllm | avishekjana | open | 0 | disc |
| Made a Demo for people who want to quickly try the model | ysong21 | open | 0 | disc |
| Compare to LightOnOCR-2-1B | staghado | open | 0 | disc |