.

app.py

@@ -11,14 +11,13 @@ from huggingface_hub import snapshot_download
 from llama_cpp import Llama
 
 repo_name = "kirp/TinyLlama-1.1B-Chat-v0.2-gguf"
-model_name = "ggml-model-q4_0.gguf"
-
+model_name = "ggml-model-q4_k_m.gguf"
 
 snapshot_download(repo_id=repo_name, local_dir=".", allow_patterns=model_name)
 
 model = Llama(
     model_path=model_name,
-    n_ctx=2048,
+    n_ctx=1024,
     n_parts=1,
 )
 
@@ -49,11 +48,11 @@ g = gr.Interface(
         gr.components.Textbox(
             lines=2, label="Prompt", value = "What is huggingface?"
         ),
-        gr.components.Slider(minimum=0, maximum=1, value=0.7, label="Temperature"),
-        gr.components.Slider(minimum=0, maximum=1, value=0.8, label="Top p"),
+        gr.components.Slider(minimum=0, maximum=1, value=0.1, label="Temperature"),
+        gr.components.Slider(minimum=0, maximum=1, value=1, label="Top p"),
         gr.components.Slider(minimum=0, maximum=100, step=1, value=50, label="Top k"),
         gr.components.Slider(
-            minimum=1, maximum=2048, step=1, value=256, label="Max tokens"
+            minimum=1, maximum=1024, step=1, value=128, label="Max tokens"
         ),
     ],
     outputs=[