Upload folder using huggingface_hub

app.py

@@ -0,0 +1,246 @@
+import gradio as gr
+import torch
+from transformers import AutoModelForCausalLM, AutoTokenizer
+import os
+
+# Model configuration
+MODEL_NAME = "tencent/HY-MT1.5-1.8B"
+
+# Global model and tokenizer instances
+tokenizer = None
+model = None
+
+def load_model():
+    """Load the model and tokenizer lazily."""
+    global tokenizer, model
+    if tokenizer is None or model is None:
+        tokenizer = AutoTokenizer.from_pretrained(MODEL_NAME)
+        model = AutoModelForCausalLM.from_pretrained(
+            MODEL_NAME,
+            device_map="auto",
+            torch_dtype=torch.bfloat16 if torch.cuda.is_available() else torch.float32
+        )
+    return tokenizer, model
+
+def generate_response(message: str, history: list, system_prompt: str = None) -> str:
+    """
+    Generate a response using the HY-MT1.5-1.8B model with chat template.
+    
+    Args:
+        message: The user's input message
+        history: List of previous conversation messages
+        system_prompt: Optional system prompt for the conversation
+        
+    Returns:
+        The model's generated response
+    """
+    try:
+        # Load model if not already loaded
+        tokenizer, model = load_model()
+        
+        # Build messages list from history
+        messages = []
+        
+        # Add system prompt if provided
+        if system_prompt:
+            messages.append({"role": "system", "content": system_prompt})
+        
+        # Add conversation history
+        for msg in history:
+            messages.append(msg)
+        
+        # Add current user message
+        messages.append({"role": "user", "content": message})
+        
+        # Apply chat template and tokenize
+        tokenized_chat = tokenizer.apply_chat_template(
+            messages,
+            tokenize=True,
+            add_generation_prompt=True,  # Add generation prompt for assistant turn
+            return_tensors="pt"
+        )
+        
+        # Generate response
+        with torch.no_grad():
+            outputs = model.generate(
+                tokenized_chat.to(model.device),
+                max_new_tokens=1024,
+                temperature=0.7,
+                top_p=0.9,
+                do_sample=True,
+                pad_token_id=tokenizer.eos_token_id
+            )
+        
+        # Decode the response
+        response = tokenizer.decode(outputs[0], skip_special_tokens=True)
+        
+        # Extract only the assistant's response (after the user's message)
+        if "assistant" in response:
+            response = response.split("assistant")[-1].strip()
+        elif "</s>" in response:
+            response = response.split("</s>")[-1].strip()
+        
+        return response
+        
+    except Exception as e:
+        return f"Error generating response: {str(e)}"
+
+def create_conversation_message(role: str, content: str) -> dict:
+    """Create a message dictionary for the conversation."""
+    return {"role": role, "content": content}
+
+# Create the Gradio 6 application
+with gr.Blocks() as demo:
+    # Application header with branding
+    gr.Markdown(
+        """
+        # 🤖 HY-MT1.5-1.8B Chatbot
+        
+        A conversational AI powered by Tencent's HY-MT1.5-1.8B model.
+        
+        ---
+        
+        **Built with** [anycoder](https://huggingface.co/spaces/akhaliq/anycoder)
+        """,
+        elem_classes=["header"]
+    )
+    
+    # Main chatbot interface
+    chat_interface = gr.ChatInterface(
+        fn=generate_response,
+        title="",
+        description="💬 Start a conversation below! The model responds to your messages using the HY-MT1.5-1.8B chat template.",
+        chatbot=gr.Chatbot(
+            placeholder="💭 How can I help you today?",
+            height=500,
+            avatar_images=(
+                "https://huggingface.co/datasets/huggingface/avatars/resolve/main/user.png",
+                "https://huggingface.co/datasets/huggingface/avatars/resolve/main/tencent.png"
+            ),
+            show_copy_all_button=True,
+            feedback_options=("👍", "👎")
+        ),
+        textbox=gr.MultimodalTextbox(
+            placeholder="Type your message here...",
+            lines=2,
+            max_lines=10,
+            submit_btn="Send ✈️",
+            stop_btn="Stop ⏹️"
+        ),
+        additional_inputs=[
+            gr.Textbox(
+                label="System Prompt (Optional)",
+                placeholder="You are a helpful assistant...",
+                lines=2,
+                max_lines=4
+            )
+        ],
+        additional_inputs_accordion=gr.Accordion(
+            label="⚙️ Advanced Settings",
+            open=False
+        ),
+        examples=[
+            ["Translate 'Hello, how are you?' into French."],
+            ["Explain quantum computing in simple terms."],
+            ["Write a short poem about the ocean."],
+            ["What are the benefits of exercise?"],
+            ["Help me plan a trip to Japan."]
+        ],
+        example_labels=["French Translation", "Quantum Computing", "Ocean Poem", "Exercise Benefits", "Japan Trip"],
+        submit_btn="Send ✈️",
+        clear_btn="Clear 🗑️",
+        autofocus=True,
+        fill_height=True,
+        api_visibility="public"
+    )
+    
+    # Model information section
+    with gr.Accordion("📋 Model Information", open=False):
+        gr.Markdown(f"""
+        ### Model Details
+        
+        - **Model**: {MODEL_NAME}
+        - **Type**: Causal Language Model with Chat Template
+        - **Provider**: [Tencent](https://huggingface.co/tencent)
+        
+        ### Capabilities
+        
+        - 📝 Text generation and completion
+        - 🌍 Translation (supports multiple languages)
+        - 💬 Conversational AI
+        - 📖 Question answering
+        - ✍️ Creative writing
+        
+        ### Usage Tips
+        
+        - Be clear and specific in your requests
+        - For translations, specify the target language
+        - Use system prompts to customize behavior
+        - Model responds in the language of your query
+        """)
+    
+    # Footer
+    gr.Markdown(
+        """
+        ---
+        
+        *This application uses the HY-MT1.5-1.8B model from Hugging Face. 
+        Responses are generated locally and are not reviewed.*
+        """,
+        elem_classes=["footer"]
+    )
+
+# Launch the application with Gradio 6 configuration
+demo.launch(
+    theme=gr.themes.Soft(
+        primary_hue="indigo",
+        secondary_hue="blue",
+        neutral_hue="slate",
+        font=gr.themes.GoogleFont("Inter"),
+        text_size="lg",
+        spacing_size="md",
+        radius_size="md"
+    ).set(
+        button_primary_background_fill="*primary_600",
+        button_primary_background_fill_hover="*primary_700",
+        block_title_text_weight="600"
+    ),
+    css="""
+    .header {
+        text-align: center;
+        padding: 20px;
+        background: linear-gradient(135deg, #667eea 0%, #764ba2 100%);
+        border-radius: 12px;
+        margin-bottom: 20px;
+    }
+    .header h1 {
+        color: white !important;
+        margin-bottom: 10px;
+    }
+    .header a {
+        color: #ffd700;
+        font-weight: bold;
+        text-decoration: none;
+    }
+    .header a:hover {
+        text-decoration: underline;
+    }
+    .footer {
+        text-align: center;
+        color: #666;
+        font-size: 0.9em;
+        padding: 10px;
+    }
+    .gradio-container {
+        max-width: 1200px !important;
+        margin: 0 auto;
+    }
+    """,
+    footer_links=[
+        {"label": "Built with anycoder", "url": "https://huggingface.co/spaces/akhaliq/anycoder"},
+        {"label": "HY-MT1.5-1.8B", "url": "https://huggingface.co/tencent/HY-MT1.5-1.8B"},
+        {"label": "Tencent", "url": "https://huggingface.co/tencent"}
+    ],
+    height=800,
+    width="100%"
+)

requirements.txt

@@ -0,0 +1,14 @@
+gradio>=6.0
+git+https://github.com/huggingface/transformers
+torch
+torchvision
+torchaudio
+accelerate
+tokenizers
+datasets
+safetensors
+sentencepiece
+Pillow
+requests
+numpy
+pandas