UI/UX improvements: language selector, dynamic reasoning, copy buttons, adjustable temperature

- Move language selector above file uploader for better visibility
- Make reasoning checkbox dynamically visible (only for Qwen3 models)
- Add copy buttons for summary and thinking panels
- Add JSON download button with metadata
- Convert temperature from locked display to adjustable slider (0.0-2.0)
- Update footer to mention bilingual support
- Add helper functions: update_reasoning_visibility, download_summary_json
- Update summarize_streaming to accept temperature parameter

app.py

@@ -35,6 +35,7 @@ AVAILABLE_MODELS = {
         "repo_id": "mradermacher/Falcon-H1-Tiny-Multilingual-100M-Instruct-GGUF",
         "filename": "*Q8_0.gguf",
         "max_context": 32768,
+        "default_temperature": 0.6,
         "inference_settings": {
             "temperature": 0.1,
             "top_p": 0.9,
@@ -47,6 +48,7 @@ AVAILABLE_MODELS = {
         "repo_id": "unsloth/gemma-3-270m-it-qat-GGUF",
         "filename": "*Q8_0.gguf",
         "max_context": 32768,
+        "default_temperature": 0.6,
         "inference_settings": {
             "temperature": 1.0,
             "top_p": 0.95,
@@ -59,6 +61,7 @@ AVAILABLE_MODELS = {
         "repo_id": "unsloth/ERNIE-4.5-0.3B-PT-GGUF",
         "filename": "*Q8_0.gguf",
         "max_context": 131072,
+        "default_temperature": 0.6,
         "inference_settings": {
             "temperature": 0.3,
             "top_p": 0.95,
@@ -71,6 +74,7 @@ AVAILABLE_MODELS = {
         "repo_id": "unsloth/granite-4.0-h-350m-GGUF",
         "filename": "*Q8_0.gguf",
         "max_context": 32768,
+        "default_temperature": 0.6,
         "inference_settings": {
             "temperature": 0.0,
             "top_p": 1.0,
@@ -83,6 +87,7 @@ AVAILABLE_MODELS = {
         "repo_id": "LiquidAI/LFM2-350M-GGUF",
         "filename": "*Q8_0.gguf",
         "max_context": 32768,
+        "default_temperature": 0.6,
         "inference_settings": {
             "temperature": 0.1,
             "top_p": 0.1,
@@ -95,6 +100,7 @@ AVAILABLE_MODELS = {
         "repo_id": "openbmb/BitCPM4-0.5B-GGUF",
         "filename": "*q4_0.gguf",
         "max_context": 131072,
+        "default_temperature": 0.6,
         "inference_settings": {
             "temperature": 0.3,
             "top_p": 0.95,
@@ -107,6 +113,7 @@ AVAILABLE_MODELS = {
         "repo_id": "mradermacher/Hunyuan-0.5B-Instruct-GGUF",
         "filename": "*Q8_0.gguf",
         "max_context": 262144,
+        "default_temperature": 0.6,
         "inference_settings": {
             "temperature": 0.3,
             "top_p": 0.95,
@@ -119,6 +126,7 @@ AVAILABLE_MODELS = {
         "repo_id": "unsloth/Qwen3-0.6B-GGUF",
         "filename": "*Q4_K_M.gguf",
         "max_context": 32768,
+        "default_temperature": 0.6,
         "supports_toggle": True,
         "inference_settings": {
             "temperature": 0.6,
@@ -132,6 +140,7 @@ AVAILABLE_MODELS = {
         "repo_id": "unsloth/Falcon-H1-1.5B-Deep-Instruct-GGUF",
         "filename": "*Q4_K_M.gguf",
         "max_context": 32768,
+        "default_temperature": 0.6,
         "inference_settings": {
             "temperature": 0.1,
             "top_p": 0.9,
@@ -144,6 +153,7 @@ AVAILABLE_MODELS = {
         "repo_id": "unsloth/Qwen3-1.7B-GGUF",
         "filename": "*Q4_K_M.gguf",
         "max_context": 32768,
+        "default_temperature": 0.6,
         "supports_toggle": True,
         "inference_settings": {
             "temperature": 0.6,
@@ -157,6 +167,7 @@ AVAILABLE_MODELS = {
         "repo_id": "mradermacher/LFM2-2.6B-Transcript-GGUF",
         "filename": "*Q4_K_M.gguf",
         "max_context": 32768,
+        "default_temperature": 0.6,
         "supports_toggle": False,
         "inference_settings": {
             "temperature": 0.3,
@@ -170,6 +181,7 @@ AVAILABLE_MODELS = {
         "repo_id": "unsloth/Qwen3-4B-Thinking-2507-GGUF",
         "filename": "*Q3_K_M.gguf",
         "max_context": 262144,
+        "default_temperature": 0.6,
         "supports_toggle": False,  # Thinking-only mode
         "inference_settings": {
             "temperature": 0.6,
@@ -183,6 +195,7 @@ AVAILABLE_MODELS = {
         "repo_id": "unsloth/granite-4.0-h-tiny-GGUF",
         "filename": "*Q3_K_M.gguf",
         "max_context": 131072,
+        "default_temperature": 0.6,
         "supports_toggle": False,
         "inference_settings": {
             "temperature": 0.0,
@@ -196,6 +209,7 @@ AVAILABLE_MODELS = {
         "repo_id": "unsloth/ERNIE-4.5-21B-A3B-Thinking-GGUF",
         "filename": "*TQ1_0.gguf",
         "max_context": 131072,
+        "default_temperature": 0.6,
         "supports_toggle": False,  # Thinking-only mode
         "inference_settings": {
             "temperature": 0.3,
@@ -270,6 +284,33 @@ def load_model(model_key: str = None) -> Tuple[Llama, str]:
         raise
 
 
+
+    def update_reasoning_visibility(model_key):
+        """Show or hide reasoning checkbox based on model capabilities."""
+        model = AVAILABLE_MODELS[model_key]
+        supports_toggle = model.get("supports_toggle", False)
+        return gr.update(visible=supports_toggle)
+
+    def download_summary_json(summary, thinking, model_key, language):
+        """Generate JSON file with summary and metadata."""
+        import json
+        from datetime import datetime
+        
+        data = {
+            "metadata": {
+                "generated_at": datetime.now().isoformat(),
+                "model": AVAILABLE_MODELS[model_key]["name"],
+                "model_id": model_key,
+                "language": language
+            },
+            "thinking_process": thinking,
+            "summary": summary
+        }
+        
+        filename = f"summary_{datetime.now().strftime('%Y%m%d_%H%M%S')}.json"
+        with open(filename, 'w', encoding='utf-8') as f:
+            json.dump(data, f, ensure_ascii=False, indent=2)
+        return filename
 def estimate_tokens(text: str) -> int:
     """
     Estimate token count for mixed CJK/English text.
@@ -378,6 +419,7 @@ def summarize_streaming(
     model_key: str,
     enable_reasoning: bool = True,
     max_tokens: int = 2048,
+    temperature: float = 0.6,
     top_p: float = None,
     top_k: int = None,
     output_language: str = "en",
@@ -729,7 +771,16 @@ def create_interface():
             # Left column - Input
             with gr.Column(scale=1):
                 with gr.Group():
-                    gr.HTML('<div class="section-header"><span class="section-icon">📤</span> Upload File</div>')
+                    gr.HTML('<div class="section-header"><span class="section-icon">🌐</span> Output Language</div>')
+                    
+                    language_selector = gr.Dropdown(
+                        choices=[("English", "en"), ("Traditional Chinese (zh-TW)", "zh-TW")],
+                        value="en",
+                        label="Select Language",
+                        info="Choose the target language for your summary"
+                    )
+                    
+                    gr.HTML('<div class="section-header" style="margin-top: 20px;"><span class="section-icon">📤</span> Upload File</div>')
                     
                     file_input = gr.File(
                         label="Drag & drop or click to upload",
@@ -746,23 +797,20 @@ def create_interface():
                                 label="Model",
                                 info="Smaller = faster. Large files need models with bigger context."
                             )
-                            language_selector = gr.Dropdown(
-                                choices=[("English", "en"), ("Traditional Chinese (zh-TW)", "zh-TW")],
-                                value="en",
-                                label="Output Language",
-                                info="Select target language for the summary"
-                            )
                             enable_reasoning = gr.Checkbox(
                                 value=True,
                                 label="Enable Reasoning Mode",
-                                info="Qwen3 only: uses /think for deeper analysis (slower) or /no_think for direct output (faster). Enabled by default.",
+                                info="Uses /think for deeper analysis (slower) or /no_think for direct output (faster). Only available for Qwen3 models.",
                                 interactive=True,
+                                visible=AVAILABLE_MODELS[DEFAULT_MODEL_KEY].get("supports_toggle", False)
                             )
-                            temperature_display = gr.Textbox(
-                                label="Temperature (Locked)",
-                                value="0.6",
-                                interactive=False,
-                                info="Set by model's recommended settings. Cannot be changed."
+                            temperature_slider = gr.Slider(
+                                minimum=0.0,
+                                maximum=2.0,
+                                value=0.6,
+                                step=0.1,
+                                label="Temperature",
+                                info="Lower = more focused, Higher = more creative"
                             )
                             max_tokens = gr.Slider(
                                 minimum=256,
@@ -824,25 +872,53 @@ def create_interface():
                         value="*Your summarized content will appear here...*",
                         elem_classes=["summary-box"]
                     )
+                    
+                    # Action buttons for outputs
+                    with gr.Row():
+                        copy_summary_btn = gr.Button("📋 Copy Summary", size="sm")
+                        copy_thinking_btn = gr.Button("📋 Copy Thinking", size="sm")
+                        download_btn = gr.Button("⬇️ Download (JSON)", size="sm")
         
         # Event handlers
         submit_btn.click(
             fn=summarize_streaming,
-            inputs=[file_input, model_dropdown, enable_reasoning, max_tokens, top_p, top_k, language_selector],
+            inputs=[file_input, model_dropdown, enable_reasoning, max_tokens, temperature_slider, top_p, top_k, language_selector],
             outputs=[thinking_output, summary_output, info_output],
             show_progress="full"
         )
         
         model_dropdown.change(
-            fn=get_model_info,
+            fn=update_reasoning_visibility,
             inputs=[model_dropdown],
-            outputs=[info_output, temperature_display, top_p, top_k],
+            outputs=[enable_reasoning]
+        )
+        
+        # Copy buttons
+        copy_summary_btn.click(
+            fn=lambda x: x,
+            inputs=[summary_output],
+            outputs=[],
+            js="(text) => { navigator.clipboard.writeText(text); return text; }"
+        )
+        
+        copy_thinking_btn.click(
+            fn=lambda x: x,
+            inputs=[thinking_output],
+            outputs=[],
+            js="(text) => { navigator.clipboard.writeText(text); return text; }"
+        )
+        
+        # Download button
+        download_btn.click(
+            fn=download_summary_json,
+            inputs=[summary_output, thinking_output, model_dropdown, language_selector],
+            outputs=[gr.File(label="Download")]
         )
         
         # Footer
         gr.HTML("""
         <div style="text-align: center; margin-top: 2rem; padding: 1rem; color: #64748b; font-size: 0.85rem; border-top: 1px solid #e2e8f0;">
-            Powered by <strong>llama-cpp-python</strong> • Running on <strong>HuggingFace Spaces Free Tier</strong><br>
+            Bilingual summaries (English & zh-TW) • Powered by <strong>llama-cpp-python</strong> • Running on <strong>HuggingFace Spaces Free Tier</strong><br>
             Traditional Chinese conversion via <strong>OpenCC</strong>
         </div>
         """)