Create app.py

app.py

@@ -0,0 +1,106 @@
+
+Copy
+
+import gradio as gr
+from transformers import pipeline, AutoTokenizer, AutoModelForCausalLM
+import torch
+ 
+# ─────────────────────────────────────────
+#  Model loading
+# ─────────────────────────────────────────
+MODEL_ID = "microsoft/phi-2"          # Change this to any HF model you like
+DEVICE   = "cuda" if torch.cuda.is_available() else "cpu"
+ 
+print(f"Loading model: {MODEL_ID}  |  Device: {DEVICE}")
+ 
+tokenizer = AutoTokenizer.from_pretrained(MODEL_ID, trust_remote_code=True)
+model     = AutoModelForCausalLM.from_pretrained(
+    MODEL_ID,
+    torch_dtype=torch.float16 if DEVICE == "cuda" else torch.float32,
+    device_map="auto",
+    trust_remote_code=True,
+)
+ 
+generator = pipeline(
+    "text-generation",
+    model=model,
+    tokenizer=tokenizer,
+    device_map="auto",
+)
+ 
+# ─────────────────────────────────────────
+#  Generation function
+# ─────────────────────────────────────────
+def generate_text(
+    prompt: str,
+    max_new_tokens: int,
+    temperature: float,
+    top_p: float,
+    top_k: int,
+    repetition_penalty: float,
+):
+    if not prompt.strip():
+        return "⚠️ Please enter a prompt."
+ 
+    outputs = generator(
+        prompt,
+        max_new_tokens=int(max_new_tokens),
+        temperature=float(temperature),
+        top_p=float(top_p),
+        top_k=int(top_k),
+        repetition_penalty=float(repetition_penalty),
+        do_sample=True,
+        pad_token_id=tokenizer.eos_token_id,
+    )
+ 
+    generated = outputs[0]["generated_text"]
+    # Remove the original prompt from the output
+    result = generated[len(prompt):].strip()
+    return result if result else "(No output generated — try a longer max_tokens)"
+ 
+# ─────────────────────────────────────────
+#  Gradio UI
+# ─────────────────────────────────────────
+with gr.Blocks(title="AI Text Generator", theme=gr.themes.Soft()) as demo:
+    gr.Markdown(
+        """
+        # 🤖 AI Text Generator
+        Powered by **Phi-2** (Microsoft) via 🤗 Transformers.
+        Enter a prompt, tweak the settings, and hit **Generate**!
+        """
+    )
+ 
+    with gr.Row():
+        with gr.Column(scale=2):
+            prompt_box = gr.Textbox(
+                label="Your Prompt",
+                placeholder="Once upon a time in a land far away...",
+                lines=5,
+            )
+            generate_btn = gr.Button("⚡ Generate", variant="primary")
+ 
+        with gr.Column(scale=1):
+            max_tokens   = gr.Slider(50, 500, value=200, step=10,  label="Max New Tokens")
+            temperature  = gr.Slider(0.1, 2.0, value=0.7, step=0.05, label="Temperature")
+            top_p        = gr.Slider(0.1, 1.0, value=0.9, step=0.05, label="Top-p")
+            top_k        = gr.Slider(1, 100, value=50, step=1,      label="Top-k")
+            rep_penalty  = gr.Slider(1.0, 2.0, value=1.1, step=0.05, label="Repetition Penalty")
+ 
+    output_box = gr.Textbox(label="Generated Text", lines=10, interactive=False)
+ 
+    gr.Examples(
+        examples=[
+            ["Write a short story about a robot learning to paint."],
+            ["Explain quantum computing in simple words:"],
+            ["Python code to reverse a linked list:\n```python\n"],
+        ],
+        inputs=prompt_box,
+    )
+ 
+    generate_btn.click(
+        fn=generate_text,
+        inputs=[prompt_box, max_tokens, temperature, top_p, top_k, rep_penalty],
+        outputs=output_box,
+    )
+ 
+demo.launch()