Update README.md

README.md

@@ -1,3 +1,78 @@
----
-license: apache-2.0
----
+license: apache-2.0
+library_name: gguf
+language:
+  - en
+pipeline_tag: text-generation
+tags:
+  - gguf
+  - llama.cpp
+  - local-llm
+  - space
+  - chat
+  - hyze
+---
+
+<p align="center">
+  <img src="https://i.imgur.com/ePJMLNp.png" alt="Hyze Logo" width="320"/>
+</p>
+
+<h1 align="center">HyzeMini (GGUF)</h1>
+
+<p align="center">
+  Lightweight GGUF builds of <b>HyzeMini</b> for fast local inference
+</p>
+
+<p align="center">
+  🔗 <a href="https://hyzeai.vercel.app">hyzeai.vercel.app</a> •  
+  📘 <a href="https://hyzedocs.vercel.app">hyzedocs.vercel.app</a> •  
+  🧠 <a href="https://hyzecode.vercel.app">hyzecode.vercel.app</a>
+</p>
+
+---
+
+## 🚀 Overview
+
+**HyzeMini (GGUF)** provides **quantized GGUF versions** of the HyzeMini model, optimized for **local execution** using tools like **llama.cpp**, **LM Studio**, **Ollama**, and other GGUF-compatible runtimes.
+
+This version keeps the same **Space + General Chat focus**, while enabling:
+
+- ⚡ Faster inference  
+- 🧠 Lower memory usage  
+- 💻 CPU-friendly execution  
+
+---
+
+## 🧠 Model Details
+
+- **Base model:** HyzeAI / HyzeMini  
+- **Parameters:** ~0.1B  
+- **Architecture:** Transformer (LLaMA-style)  
+- **Format:** GGUF  
+- **Language:** English  
+- **License:** Apache-2.0  
+
+---
+
+## 🧪 Available Quantizations
+
+*(Exact files may vary depending on upload)*
+
+Common GGUF variants include:
+
+- `Q2_K` – Ultra-low memory, fastest
+- `Q4_K_M` – Balanced quality & speed (recommended)
+- `Q5_K_M` – Higher quality, slightly slower
+- `Q8_0` – Best quality, highest memory usage
+
+> 💡 If you’re unsure, start with **Q4_K_M**.
+
+---
+
+## ⚙️ Usage
+
+### llama.cpp
+
+```bash
+./main -m HyzeMini-Q4_K_M.gguf -p "Tell me a cool space fact:"
+
+---