Update README.md

README.md

@@ -2,70 +2,74 @@
 license: mit
 language:
 - en
+pipeline_tag: text-generation
 tags:
 - gguf
 - llama.cpp
 - qwen
+- qwen2.5
 - instruct
 - lora
 - roleplay
 - holiday
 - storytelling
-pipeline_tag: text-generation
+- local-inference
+- quantization
+- ggml
+- llm
 ---
 
-# 🎀 Easter Spirit 2B (GGUF)
+# Easter Spirit 2B (GGUF)
 
-**Easter Spirit 2B** is a lightweight, seasonal language model designed for warm, cheerful, and family-friendly text generation.  
-This repository provides **GGUF** builds optimized for **local inference** using the **llama.cpp** ecosystem.
+[![License: MIT](https://img.shields.io/badge/License-MIT-green.svg)](LICENSE)
+[![Format: GGUF](https://img.shields.io/badge/Format-GGUF-orange)](#quantized-files)
+[![Runtime: llama.cpp](https://img.shields.io/badge/Runtime-llama.cpp-purple)](https://github.com/ggerganov/llama.cpp)
+[![Base: Qwen2.5-2B-Instruct](https://img.shields.io/badge/Base-Qwen2.5--2B--Instruct-lightgrey)](#model-details)
 
-🔗 **Model page:** https://huggingface.co/ghostai1/Easter-Spirit2b
+**Easter Spirit 2B** is a compact seasonal model fine-tuned for **warm, cheerful, and family-friendly** text generation.  
+This repo contains **GGUF** quantizations for **local inference** with **llama.cpp** and compatible runtimes.
+
+Model page: https://huggingface.co/ghostai1/Easter-Spirit2b
 
 ---
 
 ## Overview
 
-Easter Spirit 2B focuses on **tone, personality, and atmosphere** rather than deep reasoning or technical precision.  
-It is well suited for creative writing, holiday roleplay, and light conversational use on low-resource systems.
-
----
-
-## Model Details
+This is a **tone/personality-focused** model. It’s optimized for **wholesome, seasonal voice** and **cozy creative output**, not for deep reasoning or strict factual accuracy.
 
-- **Model name:** Easter Spirit 2B  
-- **Base model:** Qwen2.5-2B-Instruct  
-- **Fine-tuning method:** LoRA (merged)  
-- **Parameters:** ~2 billion  
-- **Format:** GGUF (llama.cpp compatible)  
-- **Language:** English  
-- **License:** MIT (base model license applies)
+Best for:
+- creative writing
+- holiday roleplay
+- light chat assistants
+- offline/local demos on modest hardware
 
-This model prioritizes:
-- Warm and friendly tone  
-- Seasonal and festive expression  
-- Wholesome, family-safe outputs  
+Not intended for:
+- complex reasoning
+- reliable factual retrieval
+- long-horizon planning
 
 ---
 
-## Intended Use
+## Model Details
 
-Recommended for:
-- Creative writing and short stories  
-- Easter and spring-themed roleplay  
-- Cozy conversational assistants  
-- Local demos and experimentation  
-- Systems with limited RAM or no GPU  
+- **Model name:** Easter Spirit 2B
+- **Base model:** Qwen2.5-2B-Instruct
+- **Fine-tuning:** LoRA (merged)
+- **Parameters:** ~2B
+- **Format:** GGUF (llama.cpp compatible)
+- **Language:** English
+- **License:** MIT (base model license applies)
 
-Not recommended for:
-- Complex reasoning tasks  
-- High-accuracy factual retrieval  
-- Long-horizon planning or analysis  
+Behavioral focus:
+- warm, friendly tone
+- spring/Easter seasonal flavor
+- family-safe outputs
 
 ---
 
 ## Quantized Files
 
-All files are generated from the same merged model and differ only in quantization level.
+All files are produced from the same merged model and differ only by quantization.
 
 | File | Quantization | Approx. Size |
 |---|---|---:|
@@ -80,18 +84,18 @@ All files are generated from the same merged model and differ only in quantizati
 | `release_v1.Q6_K.gguf` | Q6_K | ~1.27 GB |
 | `release_v1.Q8_0.gguf` | Q8_0 | ~1.65 GB |
 
-### Recommended Builds
+### Recommendations
 
-- **Default (balanced):** `Q4_K_M`  
-- **Higher quality:** `Q5_K_M`, `Q6_K`, `Q8_0`  
-- **Low-RAM systems:** `Q3_K_M`, `Q2_K`  
-- **Ultra-low memory / experimental:** `TQ1_0`  
+- **Default (balanced):** `Q4_K_M`
+- **Higher quality:** `Q5_K_M`, `Q6_K`, `Q8_0`
+- **Lower RAM:** `Q3_K_M`, `Q2_K`
+- **Ultra-low memory (experimental):** `TQ1_0`
 
 ---
 
 ## Usage (llama.cpp)
 
-### CPU-only example
+### CPU-only
 
 ```bash
 ./llama-cli \
@@ -99,3 +103,10 @@ All files are generated from the same merged model and differ only in quantizati
   -ngl 0 \
   -c 4096 \
   -p "Write a cozy springtime story inspired by Easter morning in a small town."
+
+
+./llama-cli \
+  -m release_v1.Q4_K_M.gguf \
+  -ngl 20 \
+  -c 4096 \
+  -p "Describe a cheerful Easter gathering with a warm, wholesome tone."