Upload README.md with huggingface_hub

README.md

@@ -0,0 +1,49 @@
+---
+title: JANGQ-AI
+---
+
+<p align="center">
+  <a href="https://mlx.studio"><img src="https://raw.githubusercontent.com/jjang-ai/jangq/main/assets/mlx-studio-light.png" alt="MLX Studio" width="500"></a>
+</p>
+
+<p align="center">
+  <a href="https://mlx.studio"><img src="https://mlx.studio/assets/screenshots/mlx-studio-featured.png?v=1" alt="MLX Studio App" width="600"></a>
+</p>
+
+<h3 align="center"><a href="https://mlx.studio">MLX Studio</a> — the only app that natively supports JANG models</h3>
+
+---
+
+> **LM Studio, Ollama, oMLX, Inferencer** and other MLX apps do **not** support JANG yet. Use [MLX Studio](https://mlx.studio) for native JANG support, or pip install jang for Python inference. **Ask your favorite app's creators to add JANG support!**
+
+---
+
+<p align="center">
+  <img src="https://raw.githubusercontent.com/jjang-ai/jangq/main/assets/jangq-logo-dark.png" alt="JANG" width="300">
+</p>
+
+# JANGQ-AI — JANG Quantized Models for Apple Silicon
+
+**JANG** (**J**ang **A**daptive **N**-bit **G**rading) — the GGUF equivalent for MLX.
+
+Same size as MLX, smarter bit allocation. Models stay quantized in GPU memory at full Metal speed.
+
+## Install
+
+```
+pip install "jang[mlx]"
+```
+
+## Models
+
+| Model | Profile | MMLU | HumanEval | Size |
+|-------|---------|------|-----------|------|
+| [Qwen3.5-122B-A10B-JANG_2S](https://huggingface.co/JANGQ-AI/Qwen3.5-122B-A10B-JANG_2S) | 2-bit | **84%** | **90%** | 38 GB |
+| [Qwen3.5-35B-A3B-JANG_4K](https://huggingface.co/JANGQ-AI/Qwen3.5-35B-A3B-JANG_4K) | 4-bit K-quant | **84%** | 90% | 16.7 GB |
+| [Qwen3.5-35B-A3B-JANG_2S](https://huggingface.co/JANGQ-AI/Qwen3.5-35B-A3B-JANG_2S) | 2-bit | 62% | — | 12 GB |
+
+## Links
+
+[GitHub](https://github.com/jjang-ai/jangq) · [PyPI](https://pypi.org/project/jang/) · [MLX Studio](https://mlx.studio)
+
+Created by Jinho Jang — [jangq.ai](https://jangq.ai)