Spaces:
Running
Running
Update README.md
Browse files
README.md
CHANGED
|
@@ -1,10 +1,36 @@
|
|
| 1 |
---
|
| 2 |
title: README
|
| 3 |
-
emoji:
|
| 4 |
colorFrom: indigo
|
| 5 |
colorTo: blue
|
| 6 |
sdk: static
|
| 7 |
pinned: false
|
|
|
|
|
|
|
| 8 |
---
|
| 9 |
|
| 10 |
-
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
---
|
| 2 |
title: README
|
| 3 |
+
emoji: ๐
|
| 4 |
colorFrom: indigo
|
| 5 |
colorTo: blue
|
| 6 |
sdk: static
|
| 7 |
pinned: false
|
| 8 |
+
thumbnail: >-
|
| 9 |
+
https://cdn-uploads.huggingface.co/production/uploads/67084ca7793a1fcd6c4d97ee/MxK6lKVXKh4FIp1oNRJp1.png
|
| 10 |
---
|
| 11 |
|
| 12 |
+
# Welcome to Prickly Labs ๐ต
|
| 13 |
+
|
| 14 |
+
We're a scrappy AI startup focused on training culturally-aware, emotionally realistic language models. Our lab experiments in:
|
| 15 |
+
- Low-budget, high-efficiency model training (Kaggle, FSDP, quantization)
|
| 16 |
+
- Creating unique Arabic personalities with emotional memory
|
| 17 |
+
- Discord-native inference and character interaction
|
| 18 |
+
- Dataset optimization and systematized fine-tuning tools
|
| 19 |
+
|
| 20 |
+
## ๐ง Current Projects
|
| 21 |
+
|
| 22 |
+
- **ุณูููุญ (Sanko7)**: A chaotic Egyptian bot with rubber addictions and frying pan trauma
|
| 23 |
+
- **ู
ุณุชููุน ุงูุณูุงูุญ**: A lore-based personality AI universe
|
| 24 |
+
- **Universal Notebook**: Our fine-tuning/formatting/training stack
|
| 25 |
+
- **Custom Dataset Tools**: Arabic-rich emotional data pipelines
|
| 26 |
+
|
| 27 |
+
## ๐ How to Use Our Stuff
|
| 28 |
+
|
| 29 |
+
All models will have:
|
| 30 |
+
- Chat templates
|
| 31 |
+
- Configs for `transformers`, `text-generation-inference`, and `vllm`
|
| 32 |
+
- Lightweight, Kaggle-optimized usage guides
|
| 33 |
+
|
| 34 |
+
## ๐ซ Want to Collaborate?
|
| 35 |
+
|
| 36 |
+
DM [@AhmedSherief](https://huggingface.co/ahmedasn) or reach us on Discord (coming soon).
|