Update README.md
Browse files
README.md
CHANGED
|
@@ -1,3 +1,14 @@
|
|
| 1 |
-
---
|
| 2 |
-
license: apache-2.0
|
| 3 |
-
---
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
---
|
| 2 |
+
license: apache-2.0
|
| 3 |
+
---
|
| 4 |
+
## Open Source Impact
|
| 5 |
+
|
| 6 |
+
This repository provides an MLX-optimized, 4-bit quantized DeepSeek LLM
|
| 7 |
+
for local inference on Apple silicon.
|
| 8 |
+
|
| 9 |
+
The model is publicly distributed via Hugging Face and used by members of
|
| 10 |
+
the MLX community to run on-device LLM inference without relying on cloud
|
| 11 |
+
APIs or CUDA-based stacks.
|
| 12 |
+
|
| 13 |
+
This project demonstrates practical hardware-aware ML optimization by
|
| 14 |
+
leveraging Apple’s unified memory architecture and Metal GPU acceleration.
|