DeepSeek-OCR-GGUF

DeepSeek-OCR GGUF (for llama.cpp PR #17400)

This repository provides GGUF model files required for running the DeepSeek-OCR / MTMD support introduced in the following llama.cpp pull request:

👉 llama.cpp PR #17400 https://github.com/ggml-org/llama.cpp/pull/17400

These models are only compatible with the PR branch and will not run on upstream llama.cpp main.

---

📥 Download

You can download the model files directly using:

huggingface-cli download <this-repo> --include "deepseek-ocr-bf16.gguf" --local-dir gguf_models/deepseek-ai
huggingface-cli download <this-repo> --include "mmproj-deepseek-ocr-bf16.gguf" --local-dir gguf_models/deepseek-ai

---

🚀 Run Example

Use the llama-mtmd-cli executable from the PR:

build/bin/llama-mtmd-cli \
-m gguf_models/deepseek-ai/deepseek-ocr-bf16.gguf \
--mmproj gguf_models/deepseek-ai/mmproj-deepseek-ocr-bf16.gguf \
--image tmp/mtmd_test_data/Deepseek-OCR-2510.18234v1_page1.png \
-p "<|grounding|>Convert the document to markdown." \
--chat-template deepseek-ocr --temp 0

build/bin/llama-mtmd-cli \                                                                           
-m gguf_models/deepseek-ai/deepseek-ocr-bf16_new.gguf \
--mmproj gguf_models/deepseek-ai/mmproj-deepseek-ocr-bf16_new.gguf \
--image tools/mtmd/test-1.jpeg \
-p "Free OCR. "  \
--chat-template deepseek-ocr --temp 0

build/bin/llama-server \
  -hf sabafallah/DeepSeek-OCR-GGUF:bf16 \
  --chat-template deepseek-ocr \
  --no-jinja \
  --temp 0 \
  --flash-attn off \
  --no-warmup \
  -n 4096