Update

README.md

@@ -4,7 +4,7 @@ emoji: 🏛️
 colorFrom: indigo
 colorTo: blue
 sdk: gradio
-sdk_version: 5.9.1
+sdk_version: 5.49.1
 app_file: main.py
 pinned: false
 ---

config.py

@@ -1,20 +1,20 @@
-# config.py
-
-# === API SETTINGS ===
-OPENROUTER_API_KEY = "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"
-OPENROUTER_API_URL = "https://openrouter.ai/api/v1/chat/completions"
-
-# === GOOGLE DOCS INPUT ===
-SYNTAX_DOC_URL = "https://docs.google.com/document/d/1gxQbsAIek9CMDPalQOGlCq-fmi-WRrS69_yEx6b206U/export?format=txt"
-MORPHOLOGY_DOC_URL = "https://docs.google.com/document/d/1hIal62lGso9BqfqKfB0ws8g8pBZsX-LBSTrge3M6pBM/export?format=txt"
-
-# === MODEL PRIORITY ===
-# Ordered from best to weakest. Will try top → bottom until one succeeds.
-MODEL_PRIORITY = [
-    "anthropic/claude-3-haiku",
-    "openai/gpt-3.5-turbo",
-    "nousresearch/nous-hermes-2-mistral",
-    "meta-llama/llama-3-8b-instruct",
-    "mistralai/mistral-7b-instruct",
-    "gryphe/mythomax-l2-13b"
-]
+# config.py
+
+# === API SETTINGS ===
+OPENROUTER_API_KEY = "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"
+OPENROUTER_API_URL = "https://openrouter.ai/api/v1/chat/completions"
+
+# === GOOGLE DOCS INPUT ===
+SYNTAX_DOC_URL = "https://docs.google.com/document/d/1gxQbsAIek9CMDPalQOGlCq-fmi-WRrS69_yEx6b206U/export?format=txt"
+MORPHOLOGY_DOC_URL = "https://docs.google.com/document/d/1hIal62lGso9BqfqKfB0ws8g8pBZsX-LBSTrge3M6pBM/export?format=txt"
+
+# === MODEL PRIORITY ===
+# Ordered from best to weakest. Will try top → bottom until one succeeds.
+MODEL_PRIORITY = [
+    "anthropic/claude-3-haiku",
+    "openai/gpt-3.5-turbo",
+    "nousresearch/nous-hermes-2-mistral",
+    "meta-llama/llama-3-8b-instruct",
+    "mistralai/mistral-7b-instruct",
+    "gryphe/mythomax-l2-13b"
+]

main.py

@@ -6,11 +6,6 @@ from question_runner import run_tool
 from config import MODEL_PRIORITY, SYNTAX_DOC_URL, MORPHOLOGY_DOC_URL
 from doc_utils import get_questions_from_doc
 
-# --- ZERO-GPU ENTRYPOINT (must be top-level and referenced by Gradio) ---
-def run_query(passage: str, doc_type: str):
-    # Delegate to your existing business logic
-    return run_tool(passage, doc_type)
-
 # Estimate runtime based on # of questions
 def estimate_runtime(passage, doc_type):
     if not passage or not doc_type:
@@ -22,46 +17,44 @@ def estimate_runtime(passage, doc_type):
     est_seconds = round(len(questions) * 2.5, 1)
     return f"Estimated generation time: ~{est_seconds} seconds"
 
-def build_app():
-    with gr.Blocks(theme=Soft()) as demo:
-        gr.Markdown("""
-        ## **Classical Language Query Assistant**
-        Submit a Latin or Greek passage and select the question type.
-        Answers are generated using a rotating chain of hosted AI models via OpenRouter.
-        """)
-
-        with gr.Row():
-            passage_input = gr.Textbox(label="Latin or Greek Passage", lines=4)
-            question_type = gr.Radio(["Syntax", "Morphology"], label="Question Type")
-
-        top_model = MODEL_PRIORITY[0]
-        full_model_list = "\n".join(f"- `{m}`" for m in MODEL_PRIORITY)
-        gr.Markdown(f"""
+# Build the Gradio interface at module level (required for HF Spaces)
+with gr.Blocks(theme=Soft()) as demo:
+    gr.Markdown("""
+    ## **Classical Language Query Assistant**
+    Submit a Latin or Greek passage and select the question type.
+    Answers are generated using a rotating chain of hosted AI models via OpenRouter.
+    """)
+
+    with gr.Row():
+        passage_input = gr.Textbox(label="Latin or Greek Passage", lines=4)
+        question_type = gr.Radio(["Syntax", "Morphology"], label="Question Type")
+
+    top_model = MODEL_PRIORITY[0]
+    full_model_list = "\n".join(f"- `{m}`" for m in MODEL_PRIORITY)
+    gr.Markdown(f"""
 **Currently prioritized model:** `{top_model}`  
 **Model fallback chain (if needed):**  
 {full_model_list}
 """)
 
-        with gr.Row():
-            output_text = gr.Textbox(label="Generated Answers", lines=25, interactive=False)
-            output_file = gr.File(label="Download Answers (.txt)", interactive=False)
+    estimated_time_box = gr.Textbox(label="Estimated Time", interactive=False)
 
-        estimated_time_box = gr.Textbox(label="Estimated Time", interactive=False)
+    with gr.Row():
+        output_text = gr.Textbox(label="Generated Answers", lines=25, interactive=False)
+        output_file = gr.File(label="Download Answers (.txt)", interactive=False)
 
-        passage_input.change(fn=estimate_runtime, inputs=[passage_input, question_type], outputs=estimated_time_box)
-        question_type.change(fn=estimate_runtime, inputs=[passage_input, question_type], outputs=estimated_time_box)
+    passage_input.change(fn=estimate_runtime, inputs=[passage_input, question_type], outputs=estimated_time_box)
+    question_type.change(fn=estimate_runtime, inputs=[passage_input, question_type], outputs=estimated_time_box)
 
-        submit_button = gr.Button("Generate Answers")
+    submit_button = gr.Button("Generate Answers")
 
-        # IMPORTANT: point Gradio at the decorated function
-        submit_button.click(
-            fn=run_query,
-            inputs=[passage_input, question_type],
-            outputs=[output_text, output_file, estimated_time_box],
-        )
-    return demo
+    # Connect the button to run_tool function directly
+    submit_button.click(
+        fn=run_tool,
+        inputs=[passage_input, question_type],
+        outputs=[output_text, output_file, estimated_time_box],
+        api_name="generate"
+    )
 
 if __name__ == "__main__":
-    # On Spaces: no share=True, no webbrowser.open, no infinite loop
-    app = build_app()
-    app.launch()
+    demo.launch()

question_runner.py

@@ -5,7 +5,6 @@ from router_client import query_model
 from doc_utils import get_questions_from_doc
 from config import SYNTAX_DOC_URL, MORPHOLOGY_DOC_URL
 
-# No GPU needed - we're just making API calls
 def run_tool(passage, doc_type):
     if not passage.strip():
         return "Please enter a passage to analyze.", None, None

requirements.txt

@@ -1,6 +1,6 @@
-gradio==5.49.1
-requests>=2.31.0
-spaces
-huggingface_hub>=0.20.0
---extra-index-url https://download.pytorch.org/whl/cu113
+gradio==5.49.1
+requests>=2.31.0
+spaces
+huggingface_hub>=0.20.0
+--extra-index-url https://download.pytorch.org/whl/cu113
 torch