Upload README.md with huggingface_hub

README.md

@@ -1,3 +1,23 @@
 ---
+base_model: moonshotai/Kimi-Linear-48B-A3B-Instruct
+library_name: transformers
+language:
+  - en
+tags:
+  - pytorch
 license: mit
+pipeline_tag: text-generation
+quantized_by: ymcki
 ---
+
+This is a repo for experimental GGUFs for the backend agnostic
+implementation of the Kimi-Linear model support  that requires a llama.cpp
+from this [repo](https://github.com/ymcki/llama.cpp/tree/Kimi-Linear).
+You can git clone it and compile locally.
+
+```
+git clone https://github.com/ymcki/llama.cpp --branch Kimi-Linear
+cd llama.cpp
+cmake -B build -DGGML_CUDA=ON
+cmake --build build --config Release -j 6
+```