Update app.py

app.py

@@ -17,16 +17,12 @@ subprocess.run(
     shell=True,
 )
 
-token = os.environ["HF_TOKEN"]
-
-
 model = AutoModelForCausalLM.from_pretrained(
-    "microsoft/phi-4",
-    token=token,
+    "krutrim-ai-labs/Krutrim-2-instruct",
     trust_remote_code=True,
     torch_dtype=torch.bfloat16
 )
-tok = AutoTokenizer.from_pretrained("microsoft/phi-4", token=token)
+tok = AutoTokenizer.from_pretrained("krutrim-ai-labs/Krutrim-2-instruct", token=token)
 terminators = [
     tok.eos_token_id,
 ]
@@ -80,27 +76,27 @@ def chat(message, history, temperature, do_sample, max_tokens):
 
 demo = gr.ChatInterface(
     fn=chat,
-    examples=[["Write me a poem about Machine Learning."]],
+    examples=[["Evaru Nuvvu?"]],
     # multimodal=False,
     additional_inputs_accordion=gr.Accordion(
         label="⚙️ Parameters", open=False, render=False
     ),
     additional_inputs=[
         gr.Slider(
-            minimum=0, maximum=1, step=0.1, value=0.9, label="Temperature", render=False
+            minimum=0, maximum=1, step=0.1, value=0.3, label="Temperature", render=False
         ),
         gr.Checkbox(label="Sampling", value=True),
         gr.Slider(
             minimum=128,
             maximum=4096,
             step=1,
-            value=512,
+            value=1024,
             label="Max new tokens",
             render=False,
         ),
     ],
     stop_btn="Stop Generation",
     title="Chat With LLMs",
-    description="Now Running [microsoft/phi-4](https://huggingface.co/microsoft/phi-4)",
+    description="Now Running [krutrim-ai-labs/Krutrim-2-instruct](https://huggingface.co/krutrim-ai-labs/Krutrim-2-instruct)",
 )
 demo.launch()