.

troubleshooting.md

@@ -27,9 +27,9 @@ Using older or newer versions might cause unexpected behavior with the Spaces GP
 
 ## GPU Acceleration Issues
 
-### spaces.GPU() Decorator Issues
+### spaces.GPU Decorator Issues
 
-We've observed that the `spaces.GPU()` decorator may not work correctly when used with methods inside a class. This can lead to errors like:
+We've observed that the `spaces.GPU` decorator may not work correctly when used with methods inside a class. This can lead to errors like:
 
 ```
 HTTP Request: POST http://device-api.zero/release?allowToken=... "HTTP/1.1 404 Not Found"
@@ -38,31 +38,47 @@ Error in text generation: 'GPU task aborted'
 
 ### Solution
 
-1. Use the `@spaces.GPU` decorator (without parentheses) instead of `@spaces.GPU()` with standalone functions:
+1. The syntax for spaces.GPU can be either with or without parentheses. Both of these syntaxes should work:
 
-   **Problematic:**
    ```python
-   @spaces.GPU()  # With parentheses
+   @spaces.GPU
    def generate_text(model_path, text):
        # ...
    ```
 
-   **Recommended:**
    ```python
-   @spaces.GPU  # Without parentheses
-   def generate_text_local(model_path, text):
+   @spaces.GPU()
+   def generate_text(model_path, text):
        # ...
    ```
 
-2. Use direct pipeline creation instead of loading model and tokenizer separately:
+   If you need to specify a duration for longer GPU operations, use parentheses:
+   
+   ```python
+   @spaces.GPU(duration=120)  # Set 120-second duration
+   def generate_long_text(model_path, text):
+       # ...
+   ```
+
+2. Use standalone functions instead of class methods with spaces.GPU:
 
    **Problematic:**
    ```python
-   model = AutoModelForCausalLM.from_pretrained(model_path, ...)
-   tokenizer = AutoTokenizer.from_pretrained(model_path)
-   pipe = pipeline("text-generation", model=model, tokenizer=tokenizer)
+   class ModelManager:
+       @spaces.GPU
+       def generate_text(self, model_path, text):  # Class method doesn't work well
+           # ...
+   ```
+
+   **Recommended:**
+   ```python
+   @spaces.GPU
+   def generate_text_local(model_path, text):  # Standalone function
+       # ...
    ```
 
+3. Use direct pipeline creation instead of loading model and tokenizer separately:
+
    **Recommended:**
    ```python
    tokenizer = AutoTokenizer.from_pretrained(model_path)
@@ -75,14 +91,7 @@ Error in text generation: 'GPU task aborted'
    )
    ```
 
-3. Use synchronous `InferenceClient` instead of `AsyncInferenceClient` for API calls:
-
-   **Problematic:**
-   ```python
-   from huggingface_hub import AsyncInferenceClient
-   client = AsyncInferenceClient(model_id)
-   response = await client.text_generation(text)
-   ```
+4. Use synchronous `InferenceClient` instead of `AsyncInferenceClient` for API calls:
 
    **Recommended:**
    ```python
@@ -91,7 +100,7 @@ Error in text generation: 'GPU task aborted'
    response = client.text_generation(text)  # Synchronous call
    ```
 
-4. Implement appropriate error handling to gracefully recover from GPU task aborts:
+5. Implement appropriate error handling to gracefully recover from GPU task aborts:
 
    ```python
    try: