Update app.py

app.py

@@ -4,24 +4,27 @@ import torch
 
 model_id = "deepseek-ai/deepseek-coder-7b-base"
 
-
+# Load tokenizer and model
 tokenizer = AutoTokenizer.from_pretrained(model_id)
 model = AutoModelForCausalLM.from_pretrained(
     model_id,
-    device_map="auto",          # Auto-detect GPU if available
-    torch_dtype=torch.float16   # Use FP16 for faster, lower-memory inference
+    torch_dtype=torch.float16,
+    device_map="auto"
 )
 
+# Create pipeline
 pipe = pipeline("text-generation", model=model, tokenizer=tokenizer)
 
-def generate_code(prompt):
-    response = pipe(prompt, max_new_tokens=200, temperature=0.7, do_sample=True)
-    return response[0]["generated_text"]
+# Define chat logic
+def chat(prompt):
+    output = pipe(prompt, max_new_tokens=200, do_sample=True, temperature=0.7)
+    return output[0]['generated_text']
 
+# Gradio UI
 gr.Interface(
-    fn=generate_code,
-    inputs=gr.Textbox(lines=4, placeholder="Ask DeepSeek R1 something..."),
+    fn=chat,
+    inputs=gr.Textbox(lines=2, placeholder="Ask DeepSeek 7B..."),
     outputs="text",
-    title="🧠 DeepSeek Coder R1 (1.3B)",
-    description="Running open-source DeepSeek Coder model (1.3B) on Hugging Face Spaces."
+    title="🧠 DeepSeek Coder R1 7B Chat",
+    description="7B open source code model powered by DeepSeek"
 ).launch()