diff --git a/application/llm/llama_cpp.py b/application/llm/llama_cpp.py
new file mode 100644
index 000000000..d54d6f1b2
--- /dev/null
+++ b/application/llm/llama_cpp.py
@@ -0,0 +1,35 @@
+from application.llm.base import BaseLLM
+
+class LlamaCpp(BaseLLM):
+
+    def __init__(self, api_key, llm_name='/Users/pavel/Desktop/docsgpt/application/models/orca-test.bin'):
+        global llama
+        from llama_cpp import Llama
+
+        llama = Llama(model_path=llm_name)
+
+    def gen(self, model, engine, messages, stream=False, **kwargs):
+        context = messages[0]['content']
+        user_question = messages[-1]['content']
+        prompt = f"### Instruction \n {user_question} \n ### Context \n {context} \n ### Answer \n"
+
+        result = llama(prompt, max_tokens=150, echo=False)
+
+        # import sys
+        # print(result['choices'][0]['text'].split('### Answer \n')[-1], file=sys.stderr)
+        
+        return result['choices'][0]['text'].split('### Answer \n')[-1]
+
+    def gen_stream(self, model, engine, messages, stream=True, **kwargs):
+        context = messages[0]['content']
+        user_question = messages[-1]['content']
+        prompt = f"### Instruction \n {user_question} \n ### Context \n {context} \n ### Answer \n"
+
+        result = llama(prompt, max_tokens=150, echo=False, stream=stream)
+
+        # import sys
+        # print(list(result), file=sys.stderr)
+
+        for item in result:
+            for choice in item['choices']:
+                yield choice['text']
diff --git a/application/llm/llm_creator.py b/application/llm/llm_creator.py
index a7ffc0f65..6a60f1b69 100644
--- a/application/llm/llm_creator.py
+++ b/application/llm/llm_creator.py
@@ -1,6 +1,7 @@
 from application.llm.openai import OpenAILLM, AzureOpenAILLM
 from application.llm.sagemaker import SagemakerAPILLM
 from application.llm.huggingface import HuggingFaceLLM
+from application.llm.llama_cpp import LlamaCpp
 
 
 
@@ -9,7 +10,8 @@ class LLMCreator:
         'openai': OpenAILLM,
         'azure_openai': AzureOpenAILLM,
         'sagemaker': SagemakerAPILLM,
-        'huggingface': HuggingFaceLLM
+        'huggingface': HuggingFaceLLM,
+        'llama.cpp': LlamaCpp
     }
 
     @classmethod