Add Neural Metrics card (upstream card preserved below)

README.md

@@ -10,6 +10,30 @@ tags:
 license: mit
 library_name: transformers
 ---
+<div align="center">
+
+# Neural Metrics · Compressing pages into far fewer vision tokens.
+
+<img src="https://img.shields.io/badge/Neural%20Metrics-document%20extraction-4F46E5?style=for-the-badge" alt="Neural Metrics" />
+<img src="https://img.shields.io/badge/fork%20of-deepseek--ai%2FDeepSeek--OCR-2563EB?style=flat-square" alt="fork" />
+
+</div>
+
+DeepSeek-OCR explores optical context compression - representing a full page with an order of magnitude fewer tokens than conventional pipelines. For long multi-page documents, token budget is the binding constraint, so this matters commercially.
+
+**We use it for:** long-document processing where context length is the bottleneck - cost-per-page experiments.
+
+> ### Attribution
+> This is an **unmodified fork** of [`deepseek-ai/DeepSeek-OCR`](https://huggingface.co/deepseek-ai/DeepSeek-OCR), created by the [Qwen team](https://huggingface.co/Qwen).
+> All weights, files and behaviour are identical to upstream — we rehost it so our experiments stay
+> reproducible and version-pinned. The original license and all credit remain with the Qwen team.
+> If you want the canonical model, please use [the original](https://huggingface.co/deepseek-ai/DeepSeek-OCR).
+
+---
+
+<details>
+<summary><b>Original model card from deepseek-ai/DeepSeek-OCR</b> (click to expand)</summary>
+
 <div align="center">
   <img src="https://github.com/deepseek-ai/DeepSeek-V2/blob/main/figures/logo.svg?raw=true" width="60%" alt="DeepSeek AI" />
 </div>
@@ -183,4 +207,6 @@ We also appreciate the benchmarks: [Fox](https://github.com/ucaslcl/Fox), [Omini
   author={Wei, Haoran and Sun, Yaofeng and Li, Yukun},
   journal={arXiv preprint arXiv:2510.18234},
   year={2025}
-}
+}
+
+</details>