Update app.py

app.py

@@ -8,27 +8,16 @@ import subprocess
 if not os.path.exists("ggml-model-q4_0.bin"):
     open("ggml-model-q4_0.bin", "wb").write(
         requests.get(
-            "https://huggingface.co/hlhr202/llama-7B-ggml-int4/resolve/main/ggml-model-q4_0.bin"
+            "https://huggingface.co/sasha0552/pygmalion-7b-q4_0-ggml/resolve/main/ggml-model-q4_0.bin"
         ).content
     )
-    open("tokenizer.model", "wb").write(
-        requests.get(
-            "https://huggingface.co/decapoda-research/llama-7b-hf/resolve/main/tokenizer.model"
-        ).content
-    )
-    print("Downloaded model files. Doing conversion.")
-    print(
-        subprocess.check_output(
-            "python convert.py ggml-model-q4_0.bin --outfile ggml-model.bin", shell=True
-        ).decode("utf-8")
-    )
 else:
     print("Model already exists, skipping redownload")
 
 
 print("Loading model...")
 llm = Llama(
-    model_path="ggml-model.bin",
+    model_path="ggml-model-q4_0.bin",
     seed=random.randint(1, 9999999),
     n_ctx=2048,
     n_threads=3,