Update README.md
Browse files
README.md
CHANGED
|
@@ -61,7 +61,7 @@ LLaMA is a foundational model, and as such, it should not be used for downstream
|
|
| 61 |
```
|
| 62 |
### GPU Inference in fp16
|
| 63 |
|
| 64 |
-
This requires a GPU with at least
|
| 65 |
|
| 66 |
### First, Load the Model
|
| 67 |
|
|
|
|
| 61 |
```
|
| 62 |
### GPU Inference in fp16
|
| 63 |
|
| 64 |
+
This requires a GPU with at least 15GB of VRAM.
|
| 65 |
|
| 66 |
### First, Load the Model
|
| 67 |
|