| Add MDPBench private evaluation result | Delores-Lin | open | 0 | disc |
| Multilingual powerhouse — testing for mobile deployment | 3morixd | open | 0 | disc |
| Listed on OpenModelMap | duola15 | open | 0 | disc |
| Add MDPBench evaluation results | Delores-Lin | open | 0 | disc |
| Update README.md | papafred | closed | 0 | disc |
| Переведи с сохранением структуры | Vl16161616 | open | 0 | disc |
| Add OlmOCRBench evaluation results | nielsr | open | 0 | disc |
| update README to point to new version of this model | googlefan | open | 0 | disc |
| 发票号码连续多个0的时候 会识别异常 | Alooooo | open | 0 | disc |
| ocr_test | fengshengyu | draft | 0 | disc |
| deepseekocr | sikejack | draft | 0 | disc |
| Is there a V2 in the works? | athrael-soju | open | 0 | disc |
| Can I extract the information from the invoice in JSON form using the DeepSeek OCR model? | zabih1 | open | 0 | disc |
| DeepSeek OCR Triton Error [CUDA] an illegal memory access on vLLM 0.11.2 | blazarev | open | 0 | disc |
| Deepseek-ocr cant read top right corner | Bunkid | open | 0 | disc |
| How to use Large/Gundam mode with vLLM online serving? | prudant | open | 0 | disc |
| Request: DOI | Amer9i | open | 0 | disc |
| finetune | Supperman007 | draft | 0 | disc |
| Update README.md | deleted | closed | 0 | disc |
| make deepseekocr compatible with mps and latest transformers | hebangwen | closed | 0 | disc |
| RuntimeError: The expanded size of the tensor (753) must match the existing size (752) at non-singleton dimension 0. Target sizes: [753, 1280]. Tensor sizes: [752, 1] | deathknight0 | open | 0 | disc |
| Update README.md | mrpathakji | open | 0 | disc |
| www | Matolinjo | draft | 0 | disc |
| How do I view the visualization results recognized by my DeepSeekOCR? Why is the image I saved the same as the original one? | Jiahua0 | open | 0 | disc |
| How can I disable streaming during inference? Also, can Deepseek OCR handle multiple images at the same time? | Tizzzzy | open | 0 | disc |
| Document support | wolffan | open | 0 | disc |
| [Issue] Repetitive text with endless loop | abhi22 | open | 0 | disc |
| Create Matthew Koschel The Riemann hypothesis | Mattkos | open | 0 | disc |
| OCR Request fails the first time, works second time | naamanm | open | 0 | disc |
| requirements.txt | Hifzamudassar | open | 0 | disc |
| app.py | Hifzamudassar | open | 0 | disc |
| 【问题反馈】GaoDa-M 模式下动态预处理未正确使用 1024×1024，仍固定为 640×640 问题描述 | liushihowe | open | 0 | disc |
| Lightweight FastAPI Service for DeepseekOCR + docker-compose | athrael-soju | open | 0 | disc |
| 🔥🔥🔥微调DeepSeek-OCR视频教程 | leo009 | open | 0 | disc |
| GGUF Available in NexaSDK (only). Here's how to run it locally. | alanzhuly | open | 0 | disc |
| Error while running this model in colab | dhirajsuvarna | open | 0 | disc |
| Update README.md | 16dvnk | closed | 0 | disc |
| gopu | Mauricio-100 | draft | 0 | disc |
| Request: DOI | S18069O8C2910251200058 | open | 0 | disc |
| Here is a way to self host this on AWS | mwalker714 | open | 0 | disc |
| [BUG] Missing grounding of first element in the image when the element is of type Text | MatteoOmenetti | closed | 0 | disc |
| If possible, could you inference include words with strikethrough formatting where applicable? | willy1212009 | open | 0 | disc |
| elektraspurio98@gmail.com | josiah91-me-home | open | 0 | disc |
| CPU Inferencing | chiggidev | open | 0 | disc |
| RuntimeError: The expanded size of the tensor (0) must match the existing size (1103) at non-singleton dimension 3. Target sizes: [1, 1, 1103, 0]. Tensor sizes: [1, 1, 1103, 1103] | ivanwaldon | closed | 0 | disc |
| Compare images based on scoring | schirrmacher | open | 0 | disc |
| لخص | janaismail | open | 0 | disc |
| pull | LYChuad | draft | 0 | disc |
| Update README.md | hader1890 | open | 0 | disc |
| Unrestricted conversation | von3573 | closed | 0 | disc |
| Fix: Add Transformers 4.47+ compatibility - Tested with macOS MPS | adeebaldkheel | closed | 0 | disc |
| model.infer output | Theoreticallyhugo | open | 0 | disc |
| This is what ai needed | kingriel | open | 0 | disc |
| Update README.md | DDDDARAT | open | 0 | disc |
| image_generation | Thinks123 | open | 0 | disc |
| Bboxes | mubashiross | open | 0 | disc |
| finetune ? | winzzniw555 | open | 0 | disc |
| how to use deep-seek-ocr model | Zhaleh197 | open | 0 | disc |
| How to use it correctly with online serving via vllm openai compatible server? | dhruvil237 | open | 0 | disc |
| Remove hardcoded .cuda() calls to support single forward pass on CPU and ensure DeepSeekOCR model compatibility with transformers==4.52.4 | kamalrajkannanmcw | open | 0 | disc |
| 🧠 Why does DeepSeek-OCR not use Multi-Head Latent Attention (MLA)? | ZoneTwelve | open | 0 | disc |
| Unable to execute python files | thalder-gis | closed | 0 | disc |
| Upload _cd6c77db-a9ab-4574-bc8e-bcf418df9b5c.jpg | DanniS89 | open | 0 | disc |
| How can I use this model with vLLM and send requests to it via an API? | 123fsafdfhdhfg | open | 0 | disc |
| Can i pull a best model after train for self host | pot030321 | open | 0 | disc |
| Can this model run on Ascend NPU? | Evoenzo | open | 0 | disc |
| Upload yQNWqSPsTxoEXzFZPEWAU (1).jpeg | Citrus-777 | open | 0 | disc |
| 🚩 Report: Spam | Vansh1823 | open | 0 | disc |
| Rpp | Unknown010104 | open | 0 | disc |
| Update README.md | Jalea96 | open | 0 | disc |
| Update README.md | Melskay | open | 0 | disc |
| Adding transformers as the library name | ariG23498 | closed | 0 | disc |
| ajkhjjha | djhajhk | draft | 0 | disc |
| Update README.md | krecceg | open | 0 | disc |
| Hello | anmolkumarjha | draft | 0 | disc |
| Make compatible with newer transformers | harpreetsahota | open | 0 | disc |
| Upload 3 files | kimberly-be5 | closed | 0 | disc |
| coordinates of bounding boxes along with confidences | SWAGATAM041 | open | 0 | disc |
| Genuine User Reviews and Questions on Repo deepseek-ai/DeepSeek-OCR | DeepNLP | open | 0 | disc |
| Can I ask it to do image classification? | allen7575 | open | 0 | disc |
| PDF recognition often misses several pages. | xldistance | open | 0 | disc |
| Podcast on the paper | dcaustin33 | open | 0 | disc |
| What are the minimal computer/server config (memory, processor, GPU) to run DeepSeek-OCR? | leandrobortolotto | open | 0 | disc |
| Deepseek OCR WebUI supports image and PDF file recognition. | xldistance | open | 0 | disc |
| Improve model card: add `library_name`, refine `pipeline_tag`, add HF paper and project links | nielsr | closed | 0 | disc |
| DeepSeek-OCR in DeepSeek Chat? | Ligant | open | 0 | disc |
| DeepSeek-OCR running in google colab | Javedalam | open | 0 | disc |
| 🚀测评视频 | leo009 | open | 0 | disc |
| Quick docker compose setup + vibe coded app to leverage it - very rough! | moggers | open | 0 | disc |
| add GGUF model for ollama inference | bored10 | open | 0 | disc |
| add streamer and verbose params and image_file support io.ByteIO,PIL.Image.Image for infer method | weege007 | open | 0 | disc |
| ocr | xwq777 | draft | 0 | disc |
| fixing hardcoded cuda() for cpu inference | alexgambashidze | open | 0 | disc |
| mps support | arcaputo3 | open | 0 | disc |
| 训练数据开源 | renchang | open | 0 | disc |
| Clarifying Prior Research on Visual Compression of Textual Contexts | Awiny | open | 0 | disc |
| Missing Arxiv link? | RoversX | open | 0 | disc |
| Benchmarks? | weightsnweights | open | 0 | disc |
| vllm across serving open api compatible api? | prudant | open | 0 | disc |
| Make her hands touch her lips | Sashha1 | closed | 0 | disc |
| Local Installation Video and Testing - Step by Step | fahdmirzac | open | 0 | disc |
| Model loading causes system RAM to spike and crash on Colab T4 | Sumi-AI | closed | 0 | disc |
| List of possible label types? | selimc | open | 0 | disc |
| Delete .ipynb_checkpoints/README-checkpoint.md (Duplicate, Junk) | qpqpqpqpqpqp | open | 0 | disc |
| Gradio app | freakynit | open | 0 | disc |
| Colab Demo Video and Hallucination issues | ritheshSree | open | 0 | disc |
| Flash attention issue with python 3.13 | freakynit | open | 0 | disc |
| Make infer general, so it runs on non cuda devices | aamirshakir | open | 0 | disc |
| Adding `transformers` as the library name | ariG23498 | closed | 0 | disc |
| code quality | J22 | open | 0 | disc |
| Mac support | bjai90 | open | 0 | disc |
| agi | evewashere | open | 0 | disc |
| Possibilities of QAT model versions? | Tridefender | open | 0 | disc |