Update README.md
Browse files
README.md
CHANGED
|
@@ -67,14 +67,13 @@ c. Rename the original model.onnx to other file name and put and rename the opti
|
|
| 67 |
d. Rerun step 4.
|
| 68 |
|
| 69 |
#### Speeds, Sizes, Times [optional]
|
| 70 |
-
15 token/s in Radeon 780M with 8GB
|
| 71 |
Increase to 16 token/s with device specific optimized model.onnx.<br>
|
| 72 |
As comparison, LM Studio using GGUF INT4 model and VulkanML GPU acceleration runs at 13 token/s.
|
| 73 |
|
| 74 |
#### Hardware
|
| 75 |
AMD Ryzen Zen4 7840U with integrated Radeon 780M GPU<br>
|
| 76 |
RAM 32GB<br>
|
| 77 |
-
8GB pre-allocated iGPU VRAM
|
| 78 |
|
| 79 |
#### Software
|
| 80 |
Microsoft DirectML on Windows 10
|
|
|
|
| 67 |
d. Rerun step 4.
|
| 68 |
|
| 69 |
#### Speeds, Sizes, Times [optional]
|
| 70 |
+
15 token/s in Radeon 780M with 8GB pre-allocated RAM.<br>
|
| 71 |
Increase to 16 token/s with device specific optimized model.onnx.<br>
|
| 72 |
As comparison, LM Studio using GGUF INT4 model and VulkanML GPU acceleration runs at 13 token/s.
|
| 73 |
|
| 74 |
#### Hardware
|
| 75 |
AMD Ryzen Zen4 7840U with integrated Radeon 780M GPU<br>
|
| 76 |
RAM 32GB<br>
|
|
|
|
| 77 |
|
| 78 |
#### Software
|
| 79 |
Microsoft DirectML on Windows 10
|