Update README.md
Browse files
README.md
CHANGED
|
@@ -58,6 +58,10 @@ stream = client.chat.completions.create(
|
|
| 58 |
)
|
| 59 |
```
|
| 60 |
|
|
|
|
|
|
|
|
|
|
|
|
|
| 61 |
## Benchmarks
|
| 62 |
|
| 63 |
### GPQA Diamond
|
|
|
|
| 58 |
)
|
| 59 |
```
|
| 60 |
|
| 61 |
+
### llama.cpp
|
| 62 |
+
|
| 63 |
+
Add the following command line arguments while starting llama.cpp: `--special` and `--flash-attn`.
|
| 64 |
+
|
| 65 |
## Benchmarks
|
| 66 |
|
| 67 |
### GPQA Diamond
|