v 0.1

- правка в настройках модели

app.py

@@ -42,12 +42,12 @@ llm = LlamaCPP(
     temperature=0.1,
     max_new_tokens=256,
     # llama2 has a context window of 4096 tokens, but we set it lower to allow for some wiggle room
-    context_window=16384,
+    context_window=2046,
     # kwargs to pass to __call__()
     generate_kwargs={},
     # kwargs to pass to __init__()
     # set to at least 1 to use GPU
-    model_kwargs={"n_gpu_layers": -1},
+    model_kwargs={"n_gpu_layers": -1, "num_return_sequences": 1, "no_repeat_ngram_size": 2, "n_threads": 2},
     # transform inputs into Llama2 format
     messages_to_prompt=messages_to_prompt,
     completion_to_prompt=completion_to_prompt,