Update README.md
Browse files
README.md
CHANGED
|
@@ -25,28 +25,33 @@ tags:
|
|
| 25 |
[](#quantized-files)
|
| 26 |
[](https://github.com/ggerganov/llama.cpp)
|
| 27 |
[](#model-details)
|
|
|
|
|
|
|
| 28 |
|
| 29 |
**Easter Spirit 2B** is a compact seasonal model fine-tuned for **warm, cheerful, and family-friendly** text generation.
|
| 30 |
-
This
|
| 31 |
|
| 32 |
-
|
|
|
|
|
|
|
|
|
|
| 33 |
|
| 34 |
---
|
| 35 |
|
| 36 |
## Overview
|
| 37 |
|
| 38 |
-
This is a **tone/personality-focused** model. It
|
| 39 |
|
| 40 |
-
|
| 41 |
-
-
|
| 42 |
-
-
|
| 43 |
-
-
|
| 44 |
-
-
|
| 45 |
|
| 46 |
-
Not
|
| 47 |
-
-
|
| 48 |
-
-
|
| 49 |
-
-
|
| 50 |
|
| 51 |
---
|
| 52 |
|
|
@@ -60,16 +65,11 @@ Not intended for:
|
|
| 60 |
- **Language:** English
|
| 61 |
- **License:** MIT (base model license applies)
|
| 62 |
|
| 63 |
-
Behavioral focus:
|
| 64 |
-
- warm, friendly tone
|
| 65 |
-
- spring/Easter seasonal flavor
|
| 66 |
-
- family-safe outputs
|
| 67 |
-
|
| 68 |
---
|
| 69 |
|
| 70 |
## Quantized Files
|
| 71 |
|
| 72 |
-
All files are produced from the same merged model and differ only
|
| 73 |
|
| 74 |
| File | Quantization | Approx. Size |
|
| 75 |
|---|---|---:|
|
|
@@ -88,7 +88,7 @@ All files are produced from the same merged model and differ only by quantizatio
|
|
| 88 |
|
| 89 |
- **Default (balanced):** `Q4_K_M`
|
| 90 |
- **Higher quality:** `Q5_K_M`, `Q6_K`, `Q8_0`
|
| 91 |
-
- **
|
| 92 |
- **Ultra-low memory (experimental):** `TQ1_0`
|
| 93 |
|
| 94 |
---
|
|
@@ -103,10 +103,3 @@ All files are produced from the same merged model and differ only by quantizatio
|
|
| 103 |
-ngl 0 \
|
| 104 |
-c 4096 \
|
| 105 |
-p "Write a cozy springtime story inspired by Easter morning in a small town."
|
| 106 |
-
|
| 107 |
-
|
| 108 |
-
./llama-cli \
|
| 109 |
-
-m release_v1.Q4_K_M.gguf \
|
| 110 |
-
-ngl 20 \
|
| 111 |
-
-c 4096 \
|
| 112 |
-
-p "Describe a cheerful Easter gathering with a warm, wholesome tone."
|
|
|
|
| 25 |
[](#quantized-files)
|
| 26 |
[](https://github.com/ggerganov/llama.cpp)
|
| 27 |
[](#model-details)
|
| 28 |
+
[](https://huggingface.co/ghostai1/Easter-Spirit2b)
|
| 29 |
+
[](https://www.linkedin.com/in/ccengineering/)
|
| 30 |
|
| 31 |
**Easter Spirit 2B** is a compact seasonal model fine-tuned for **warm, cheerful, and family-friendly** text generation.
|
| 32 |
+
This repository provides **GGUF** builds optimized for **local inference** using the **llama.cpp** ecosystem and compatible runtimes.
|
| 33 |
|
| 34 |
+
**Quick links**
|
| 35 |
+
- Model: https://huggingface.co/ghostai1/Easter-Spirit2b
|
| 36 |
+
- Runtime: https://github.com/ggerganov/llama.cpp
|
| 37 |
+
- Author (LinkedIn): https://www.linkedin.com/in/ccengineering/
|
| 38 |
|
| 39 |
---
|
| 40 |
|
| 41 |
## Overview
|
| 42 |
|
| 43 |
+
This is a **tone/personality-focused** model. It emphasizes warmth, friendliness, and seasonal flavor rather than deep reasoning or strict technical accuracy.
|
| 44 |
|
| 45 |
+
Recommended for:
|
| 46 |
+
- Creative writing and short stories
|
| 47 |
+
- Holiday / spring-themed roleplay
|
| 48 |
+
- Light conversational assistants
|
| 49 |
+
- Local demos and low-resource systems
|
| 50 |
|
| 51 |
+
Not optimized for:
|
| 52 |
+
- Complex reasoning
|
| 53 |
+
- Factual retrieval
|
| 54 |
+
- Long-horizon planning
|
| 55 |
|
| 56 |
---
|
| 57 |
|
|
|
|
| 65 |
- **Language:** English
|
| 66 |
- **License:** MIT (base model license applies)
|
| 67 |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 68 |
---
|
| 69 |
|
| 70 |
## Quantized Files
|
| 71 |
|
| 72 |
+
All files are produced from the same merged model and differ only in quantization.
|
| 73 |
|
| 74 |
| File | Quantization | Approx. Size |
|
| 75 |
|---|---|---:|
|
|
|
|
| 88 |
|
| 89 |
- **Default (balanced):** `Q4_K_M`
|
| 90 |
- **Higher quality:** `Q5_K_M`, `Q6_K`, `Q8_0`
|
| 91 |
+
- **Low RAM systems:** `Q3_K_M`, `Q2_K`
|
| 92 |
- **Ultra-low memory (experimental):** `TQ1_0`
|
| 93 |
|
| 94 |
---
|
|
|
|
| 103 |
-ngl 0 \
|
| 104 |
-c 4096 \
|
| 105 |
-p "Write a cozy springtime story inspired by Easter morning in a small town."
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|