Update README.md

README.md

@@ -1,74 +1,64 @@
 ---
 license: mit
 ---
----
-language:
-- en
-tags:
-- gguf
-- llama.cpp
-- mistral
-- finetune
-- lora
-- roleplay
-- instruct
-license: apache-2.0
-base_model: mistralai/Mistral-7B-Instruct-v0.3
-library_name: llama.cpp
-pipeline_tag: text-generation
----
 
-# GHOSTAI — Spooky Mistral (GGUF Release)
+<!--
+GHOSTAI • HORROR GGUF RELEASE README
+Drop this into README.md at the root of your Hugging Face repo.
+-->
 
-This is the **GGUF release** of **GHOSTAI**: a fine-tuned variant of **Mistral-7B-Instruct v0.3** exported for **llama.cpp**-compatible runtimes.
+<p align="center">
+  <img src="https://capsule-render.vercel.app/api?type=waving&color=0:0b0b0f,50:2b0a2a,100:0b0b0f&height=160&section=header&text=GHOSTAI%20%E2%80%94%20HORROR%20GGUF&fontSize=44&fontColor=EAEAEA&animation=twinkling" />
+</p>
 
-It ships multiple quantizations (with spooky filenames) so you can pick the right tradeoff for quality vs memory.
+<p align="center">
+  <img alt="GGUF" src="https://img.shields.io/badge/GGUF-llama.cpp-8A2BE2?style=for-the-badge">
+  <img alt="Base" src="https://img.shields.io/badge/Base-Mistral%207B%20Instruct%20v0.3-5B2C83?style=for-the-badge">
+  <img alt="Quant" src="https://img.shields.io/badge/Quant-Q4__K__M%20%7C%20F16-3A0CA3?style=for-the-badge">
+  <img alt="Theme" src="https://img.shields.io/badge/Theme-Horror-8B0000?style=for-the-badge">
+</p>
 
-## What’s in this repo
+<p align="center">
+  <b>GHOSTAI</b> is a horror-flavored GGUF release (llama.cpp-ready) built from a LoRA fine-tune on <code>mistralai/Mistral-7B-Instruct-v0.3</code>.
+  <br/>
+  Pick your haunt: <b>F16</b> for max fidelity or <b>Q4_K_M</b> for the best everyday balance.
+</p>
 
-You’ll find:
+---
 
-- A **full-precision GGUF** (f16) export
-- One or more **quantized GGUF** files for fast inference
+## 🩸 What’s inside
 
-### Spooky GGUF names (file mapping)
+This repo contains **GGUF** files for fast local inference using **llama.cpp**-compatible runtimes.
 
-These are just friendly names; the actual quant format is in the table.
+### 🎃 Spooky file set
 
-| Spooky name | Actual file | Quant format | Notes |
+| Codename | File | Format | Use case |
 |---|---|---:|---|
-| **GHOSTAI_FOGF16** | `model.f16.gguf` | f16 | Highest quality; largest file |
-| **GHOSTAI_CRYPT_Q4KM** | `model.Q4_K_M.gguf` | Q4_K_M | Recommended default balance |
-| **GHOSTAI_WHISPER_IQ1S** | `model.IQ1_S.gguf` | IQ1_S | Extremely small; quality drop |
-| *(fallback)* **GHOSTAI_RAGDOLL_Q2K** | `model.Q2_K.gguf` | Q2_K | If IQ1_S not supported |
-
-If a “spooky” file name isn’t present, that quant was not generated for this release.
-
-## Base model
+| **GHOSTAI_FOGF16** | `model.f16.gguf` | f16 | Maximum quality (largest) |
+| **GHOSTAI_CRYPT_Q4KM** | `model.Q4_K_M.gguf` | Q4_K_M | Best default (quality/size) |
+| **GHOSTAI_WHISPER_IQ1S** | `model.IQ1_S.gguf` | IQ1_S | Tiny build (quality drop) |
+| **GHOSTAI_RAGDOLL_Q2K** | `model.Q2_K.gguf` | Q2_K | Fallback if IQ1_S unsupported |
 
-- Base: `mistralai/Mistral-7B-Instruct-v0.3`
+> Not all files may exist in every release—this table lists the intended set. Use the “Files” panel to confirm what’s included.
 
-## Training summary
-
-- Method: LoRA fine-tune
-- Export pipeline: LoRA adapter → merged weights → GGUF conversion → quantization
-- Intended use: general instruction/chat (custom domain depends on your dataset)
-
-> If you want to disclose dataset or alignment details, add them here (recommended).
+---
 
-## Quickstart (llama.cpp)
+## 🧬 Base model
 
-### 1) Download a GGUF
+- **Base**: `mistralai/Mistral-7B-Instruct-v0.3`
+- **Release type**: GGUF export (llama.cpp ecosystem)
+- **Training method**: LoRA fine-tune → merged → GGUF → quantized
 
-Pick one file (start with `model.Q4_K_M.gguf` if you’re unsure).
+---
 
-### 2) Run with llama.cpp
+## ⚰️ Quickstart (llama.cpp)
 
-Example using `llama-cli`:
+### 1) Run on GPU (CUDA build)
 
 ```bash
 ./llama-cli \
   -m model.Q4_K_M.gguf \
   -ngl 99 \
   -c 4096 \
-  -p "You are GHOSTAI. Introduce yourself in one paragraph."
+  -p "You are GHOSTAI. Speak like a calm narrator in a horror novel. Keep it concise."
+