Update README.md
Browse files
README.md
CHANGED
|
@@ -5,6 +5,36 @@ colorFrom: pink
|
|
| 5 |
colorTo: purple
|
| 6 |
sdk: static
|
| 7 |
pinned: false
|
|
|
|
| 8 |
---
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 9 |
|
| 10 |
-
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 5 |
colorTo: purple
|
| 6 |
sdk: static
|
| 7 |
pinned: false
|
| 8 |
+
license: mit
|
| 9 |
---
|
| 10 |
+
<div align="center">
|
| 11 |
+
<h1>🚀 CQ Systems</h1>
|
| 12 |
+
<h3><i>To Support The Democratization of Foundational Intelligence</i></h3>
|
| 13 |
+
<a href="https://www.cqurity.ai" target="_blank">www.cqurity.ai</a>
|
| 14 |
+
</div>
|
| 15 |
|
| 16 |
+
---
|
| 17 |
+
|
| 18 |
+
## About Us
|
| 19 |
+
At **CQ Systems**, we are deeply invested in the advancement of Artificial Intelligence and Machine Learning. Our primary mission is to democratize foundational intelligence by optimizing state-of-the-art open-weights models for consumer hardware, edge devices, and advanced silicon architectures.
|
| 20 |
+
|
| 21 |
+
We believe that frontier-level AI should be accessible, efficient, and ready to run locally.
|
| 22 |
+
|
| 23 |
+
## Our Work
|
| 24 |
+
We specialize in highly optimized **GGUF quantizations** of top-tier foundation models. Our current focus is on the Google Gemma 4 family, providing researchers and developers with a spectrum of models that balance extreme memory efficiency with uncompromising reasoning capabilities.
|
| 25 |
+
|
| 26 |
+
### 🌟 Featured Models
|
| 27 |
+
|
| 28 |
+
| Model | Size / Type | Description |
|
| 29 |
+
| :--- | :--- | :--- |
|
| 30 |
+
| [**Google-Gemma-E4B-IT**](https://huggingface.co/CQSystems/Google-Gemma-E4B-IT) | 8B (4.5B Active) | *Our most popular release.* A highly efficient, instruction-tuned model perfect for edge devices and Apple Silicon. Features advanced iMatrix quantizations. |
|
| 31 |
+
| [**Google-Gemma-4-31B-It**](https://huggingface.co/CQSystems/Google-Gemma-4-31B-It) | 31B Dense | The heavyweight flagship of the Gemma 4 family, heavily compressed for local execution without sacrificing frontier-level reasoning. |
|
| 32 |
+
| [**Google-Gemma-4-E2B-It**](https://huggingface.co/CQSystems/Google-Gemma-4-E2B-It) | 5B (2.3B Active) | Ultra-lightweight and lightning fast. Purpose-built for maximum performance on constrained memory systems. |
|
| 33 |
+
|
| 34 |
+
---
|
| 35 |
+
|
| 36 |
+
## Connect with Us
|
| 37 |
+
* 🌐 **Website:** [cqurity.ai](https://www.cqurity.ai)
|
| 38 |
+
* 💡 **Interests:** AI, Machine Learning, Quantization, Local LLMs, Model Optimization
|
| 39 |
+
|
| 40 |
+
*Building the future of local AI, one optimized tensor at a time.*
|