Commit
•
f542954
1
Parent(s):
2223e17
Move back to 8bit quantization
Browse files
app.py
CHANGED
|
@@ -11,8 +11,8 @@ logging.basicConfig(level=logging.INFO)
|
|
| 11 |
|
| 12 |
|
| 13 |
model_repo="filipealmeida/open-llama-3b-v2-pii-transform"
|
| 14 |
-
model_filename="ggml-model-f16.gguf"
|
| 15 |
-
|
| 16 |
|
| 17 |
def download_model():
|
| 18 |
print("Downloading model...")
|
|
|
|
| 11 |
|
| 12 |
|
| 13 |
model_repo="filipealmeida/open-llama-3b-v2-pii-transform"
|
| 14 |
+
# model_filename="ggml-model-f16.gguf"
|
| 15 |
+
model_filename="ggml-model-Q8_0.gguf"
|
| 16 |
|
| 17 |
def download_model():
|
| 18 |
print("Downloading model...")
|