Update README.md
Browse files
README.md
CHANGED
@@ -11,7 +11,7 @@ Due to the license regulation of LLaMA, we are not allowed to release the accele
|
|
11 |
|
12 |
## Model Card for lyraLLaMA
|
13 |
|
14 |
-
lyraLLaMA is currently the **fastest LLaMA-13b** available. The inference speed of lyraLLaMA has achieved **4x** acceleration upon the torch version.
|
15 |
|
16 |
Among its main features are:
|
17 |
- device: Nvidia GPU with Amperer architecture or Volta architecture (A100 or higher, V100).
|
|
|
11 |
|
12 |
## Model Card for lyraLLaMA
|
13 |
|
14 |
+
lyraLLaMA is currently the **fastest LLaMA-13b** available. The inference speed of lyraLLaMA has achieved **3000+ tokens/s** on A100, **4x** acceleration upon the torch version.
|
15 |
|
16 |
Among its main features are:
|
17 |
- device: Nvidia GPU with Amperer architecture or Volta architecture (A100 or higher, V100).
|