Update app.py
Browse filesLower max_new_tokens to 250. It still takes an eternity to respond with the 135M model.
app.py
CHANGED
|
@@ -4,7 +4,7 @@ import gradio as gr
|
|
| 4 |
from transformers import pipeline
|
| 5 |
import torch
|
| 6 |
|
| 7 |
-
MAX_NEW_TOKENS =
|
| 8 |
|
| 9 |
MODEL="HuggingFaceTB/SmolLM2-135M-Instruct"
|
| 10 |
# MODEL="HuggingFaceTB/SmolLM2-360M-Instruct"
|
|
|
|
| 4 |
from transformers import pipeline
|
| 5 |
import torch
|
| 6 |
|
| 7 |
+
MAX_NEW_TOKENS = 250
|
| 8 |
|
| 9 |
MODEL="HuggingFaceTB/SmolLM2-135M-Instruct"
|
| 10 |
# MODEL="HuggingFaceTB/SmolLM2-360M-Instruct"
|