Spaces:
Running
on
Zero
Running
on
Zero
Update app.py
Browse files
app.py
CHANGED
@@ -51,7 +51,7 @@ current_model = None
|
|
51 |
current_tokenizer = None
|
52 |
current_model_path = None
|
53 |
|
54 |
-
@spaces.GPU(duration=
|
55 |
def load_model(model_path, progress=gr.Progress()):
|
56 |
"""Load the selected model and tokenizer"""
|
57 |
global current_model, current_tokenizer, current_model_path
|
@@ -86,7 +86,7 @@ def load_model(model_path, progress=gr.Progress()):
|
|
86 |
progress(1.0, desc="Model loading failed!")
|
87 |
return f"Model loading failed: {str(e)}"
|
88 |
|
89 |
-
@spaces.GPU(duration=
|
90 |
def generate_response_non_streaming(instruction, model_name, temperature=0.7, max_tokens=1024):
|
91 |
"""Generate a response from the Apollo model (non-streaming)"""
|
92 |
global current_model, current_tokenizer, current_model_path
|
|
|
51 |
current_tokenizer = None
|
52 |
current_model_path = None
|
53 |
|
54 |
+
@spaces.GPU(duration=120)
|
55 |
def load_model(model_path, progress=gr.Progress()):
|
56 |
"""Load the selected model and tokenizer"""
|
57 |
global current_model, current_tokenizer, current_model_path
|
|
|
86 |
progress(1.0, desc="Model loading failed!")
|
87 |
return f"Model loading failed: {str(e)}"
|
88 |
|
89 |
+
@spaces.GPU(duration=120)
|
90 |
def generate_response_non_streaming(instruction, model_name, temperature=0.7, max_tokens=1024):
|
91 |
"""Generate a response from the Apollo model (non-streaming)"""
|
92 |
global current_model, current_tokenizer, current_model_path
|