Update API documentation to clarify Hugging Face Spaces queue requirements

CLAUDE.md

@@ -36,22 +36,21 @@ The application consists of a single `app.py` file with:
 
 ## Important Configuration Details
 
-- **No Queue**: The app launches without `.queue()` to allow direct API calls
+- **Queue**: Hugging Face Spaces enforces queuing at infrastructure level, even without `.queue()` in code
 - **CPU Mode**: Explicitly set to CPU to avoid GPU requirements
 - **Trust Remote Code**: Model uses `trust_remote_code=True` for custom Nomic model code
-- **API Structure**: Uses `fn_index: 0` for the predict endpoint
+- **API Access**: Direct HTTP requires queue protocol; use Gradio client libraries instead
 
 ## API Usage
 
-The `/api/predict` endpoint expects:
-```json
-{
-  "fn_index": 0,
-  "data": ["text to embed"]
-}
+Use Gradio client libraries for API access:
+```python
+from gradio_client import Client
+client = Client("ipepe/nomic-embeddings")
+result = client.predict("text to embed", api_name="/predict")
 ```
 
-Returns embedding array in response.data field.
+Direct HTTP requires implementing Gradio's queue protocol (join queue, SSE listening, session management).
 
 ## Deployment Notes
 

app.py

@@ -30,30 +30,18 @@ with gr.Blocks(title="Nomic Text Embeddings") as app:
     # Add API usage guide
     gr.Markdown("## API Usage")
     gr.Markdown("""
-    You can use this API programmatically with direct HTTP requests (queue disabled).
+    You can use this API programmatically. Hugging Face Spaces requires using their client libraries which handle queuing automatically.
     
-    ### Using cURL
+    ### Quick Command-Line Usage
     ```bash
-    curl -X POST https://ipepe-nomic-embeddings.hf.space/api/predict \
-      -H "Content-Type: application/json" \
-      -d '{
-        "fn_index": 0,
-        "data": ["Your text to embed goes here"]
-      }'
-    ```
-    
-    ### Python Example (using requests)
-    ```python
-    import requests
+    # Install gradio client
+    pip install gradio_client
     
-    response = requests.post(
-        "https://ipepe-nomic-embeddings.hf.space/api/predict",
-        json={"fn_index": 0, "data": ["Your text to embed goes here"]}
-    )
-    print(response.json())  # Returns the embedding array
+    # Generate embedding with one command
+    python -c "from gradio_client import Client; print(Client('ipepe/nomic-embeddings').predict('Your text here', api_name='/predict'))"
     ```
     
-    ### Python Example (using Gradio Client)
+    ### Python Example (Recommended)
     ```python
     from gradio_client import Client
     
@@ -74,6 +62,14 @@ with gr.Blocks(title="Nomic Text Embeddings") as app:
     console.log(result.data);
     ```
     
+    ### Direct HTTP (Advanced)
+    Direct HTTP requests require implementing the Gradio queue protocol:
+    1. POST to `/queue/join` to join queue
+    2. Listen to `/queue/data` via SSE for results
+    3. Handle session management
+    
+    For direct HTTP, we recommend using the official Gradio clients above which handle this automatically.
+    
     The response will contain the embedding array as a list of floats.
     """)