Create app.py

app.py

@@ -0,0 +1,43 @@
+import torch
+from transformers import AutoTokenizer, AutoModelForCausalLM   # or your model class
+import gradio as gr
+
+# load tokenizer & model
+model_name = "FlameF0X/i3-80m"   # replace with correct HF model path
+tokenizer = AutoTokenizer.from_pretrained(model_name)
+model = AutoModelForCausalLM.from_pretrained(model_name)
+model.eval()
+if torch.cuda.is_available():
+    model = model.cuda()
+
+def generate(prompt: str, max_new_tokens: int = 100, temperature: float = 1.0, top_k: int = None):
+    inputs = tokenizer(prompt, return_tensors="pt")
+    input_ids = inputs["input_ids"]
+    if torch.cuda.is_available():
+        input_ids = input_ids.cuda()
+    with torch.no_grad():
+        output_ids = model.generate(
+            input_ids,
+            max_new_tokens=max_new_tokens,
+            temperature=temperature,
+            top_k=top_k,
+            do_sample=True
+        )
+    output = tokenizer.decode(output_ids[0], skip_special_tokens=True)
+    return output
+
+# Gradio interface
+iface = gr.Interface(
+    fn=generate,
+    inputs=[
+        gr.Textbox(label="Prompt", lines=2, placeholder="Enter prompt here..."),
+        gr.Slider(label="Max new tokens", minimum=1, maximum=500, step=1, value=100),
+        gr.Slider(label="Temperature", minimum=0.1, maximum=2.0, step=0.1, value=1.0),
+        gr.Slider(label="Top-k (0 = disabled)", minimum=0, maximum=200, step=1, value=40)
+    ],
+    outputs=gr.Textbox(label="Generated Text"),
+    title="i3-80m Generation Demo",
+    description="Interact with the i3 hybrid-architecture model."
+)
+
+iface.launch(server_name="0.0.0.0", server_port=7860)