diff --git a/CONTRIBUTORS.md b/CONTRIBUTORS.md
index 00cf16f06..297b91930 100644
--- a/CONTRIBUTORS.md
+++ b/CONTRIBUTORS.md
@@ -8,6 +8,7 @@ Thanks to these individuals for making reNgine awesome by fixing bugs, resolving
 * [Suprita-25](https://github.com/Suprita-25)
 * [TheBinitGhimire](https://github.com/TheBinitGhimire)
 * [Vinay Leo](https://github.com/vinaynm)
+* [Erdem Ozgen](https://github.com/ErdemOzgen)
 
 *If you have created a Pull request, feel free to add your name here, because we know you are awesome and deserve thanks from the community!*
 
diff --git a/Makefile b/Makefile
index f1b8d1d1d..e37b2af73 100644
--- a/Makefile
+++ b/Makefile
@@ -6,7 +6,7 @@
 COMPOSE_PREFIX_CMD := COMPOSE_DOCKER_CLI_BUILD=1
 
 COMPOSE_ALL_FILES := -f docker-compose.yml
-SERVICES          := db web proxy redis celery celery-beat
+SERVICES          := db web proxy redis celery celery-beat ollama
 
 # --------------------------
 
@@ -20,7 +20,7 @@ setup:			## Generate certificates.
 
 up:				## Build and start all services.
 	${COMPOSE_PREFIX_CMD} docker-compose ${COMPOSE_ALL_FILES} up -d --build ${SERVICES}
-
+	docker exec -it ollama ollama run llama2
 build:			## Build all services.
 	${COMPOSE_PREFIX_CMD} docker-compose ${COMPOSE_ALL_FILES} build ${SERVICES}
 
diff --git a/docker-compose.dev.yml b/docker-compose.dev.yml
index 4fbe8e539..496f210b9 100644
--- a/docker-compose.dev.yml
+++ b/docker-compose.dev.yml
@@ -115,6 +115,17 @@ services:
       - celery-beat
     networks:
       - rengine_network
+  ollama:
+    image: ollama/ollama
+    container_name: ollama
+    volumes:
+      - ollama_data:/root/.ollama
+    ports:
+      - "11434:11434"
+    networks:
+      - rengine_network
+    restart: always
+    command: ["ollama", "run", "llama2-uncensored"]
 
 networks:
   rengine_network:
@@ -126,3 +137,4 @@ volumes:
   github_repos:
   wordlist:
   scan_results:
+  ollama_data:
diff --git a/docker-compose.yml b/docker-compose.yml
index 9bd3e018a..d575104ae 100644
--- a/docker-compose.yml
+++ b/docker-compose.yml
@@ -140,7 +140,15 @@ services:
       - scan_results:/usr/src/scan_results
     networks:
       - rengine_network
-
+  ollama:
+    image: ollama/ollama
+    container_name: ollama
+    volumes:
+      - ollama_data:/root/.ollama
+    ports:
+      - "11434:11434"
+    networks:
+      - rengine_network
 
 networks:
   rengine_network:
@@ -154,6 +162,7 @@ volumes:
   wordlist:
   scan_results:
   static_volume:
+  ollama_data:
 
 secrets:
   proxy.ca:
diff --git a/web/celery-entrypoint.sh b/web/celery-entrypoint.sh
index f1d49ff1d..775fd1869 100755
--- a/web/celery-entrypoint.sh
+++ b/web/celery-entrypoint.sh
@@ -159,6 +159,8 @@ exec "$@"
 # httpx seems to have issue, use alias instead!!!
 echo 'alias httpx="/go/bin/httpx"' >> ~/.bashrc
 
+# for localgpt install langchain
+python3 -m pip install langchain==0.0.343
 
 # watchmedo auto-restart --recursive --pattern="*.py" --directory="/usr/src/app/reNgine/" -- celery -A reNgine.tasks worker --autoscale=10,0 -l INFO -Q scan_queue &
 echo "Starting Workers..."
diff --git a/web/reNgine/gpt.py b/web/reNgine/gpt.py
index 4bae2563e..2e6682ff1 100644
--- a/web/reNgine/gpt.py
+++ b/web/reNgine/gpt.py
@@ -2,13 +2,16 @@
 import re
 from reNgine.common_func import get_open_ai_key, extract_between
 from reNgine.definitions import VULNERABILITY_DESCRIPTION_SYSTEM_MESSAGE, ATTACK_SUGGESTION_GPT_SYSTEM_PROMPT
+from langchain.llms import Ollama
 
 class GPTVulnerabilityReportGenerator:
 
 	def __init__(self):
 		self.api_key = get_open_ai_key()
 		self.model_name = 'gpt-3.5-turbo'
-
+		if not self.api_key:
+			self.ollama = Ollama(base_url='http://ollama:11434', model="llama2-uncensored")
+	
 	def get_vulnerability_description(self, description):
 		"""Generate Vulnerability Description using GPT.
 
@@ -24,93 +27,91 @@ def get_vulnerability_description(self, description):
 			}
 		"""
 		if not self.api_key:
-			return {
-				'status': False,
-				'error': 'No OpenAI keys provided.'
-			}
-		openai.api_key = self.api_key
-		try:
-			gpt_response = openai.ChatCompletion.create(
-			model=self.model_name,
-			messages=[
-					{'role': 'system', 'content': VULNERABILITY_DESCRIPTION_SYSTEM_MESSAGE},
-					{'role': 'user', 'content': description}
-				]
-			)
+			prompt = ATTACK_SUGGESTION_GPT_SYSTEM_PROMPT + "\nUser: " + input
+			response_content = self.ollama(prompt)
+		else:
+			openai.api_key = self.api_key
+			try:
+				gpt_response = openai.ChatCompletion.create(
+				model=self.model_name,
+				messages=[
+						{'role': 'system', 'content': VULNERABILITY_DESCRIPTION_SYSTEM_MESSAGE},
+						{'role': 'user', 'content': description}
+					]
+				)
 
-			response_content = gpt_response['choices'][0]['message']['content']
+				response_content = gpt_response['choices'][0]['message']['content']
+			except Exception as e:
+				return {
+					'status': False,
+					'error': str(e)
+				}
+		vuln_description_pattern = re.compile(
+			r"[Vv]ulnerability [Dd]escription:(.*?)(?:\n\n[Ii]mpact:|$)",
+			re.DOTALL
+		)
+		impact_pattern = re.compile(
+			r"[Ii]mpact:(.*?)(?:\n\n[Rr]emediation:|$)",
+			re.DOTALL
+		)
+		remediation_pattern = re.compile(
+			r"[Rr]emediation:(.*?)(?:\n\n[Rr]eferences:|$)",
+			re.DOTALL
+		)
 
-			vuln_description_pattern = re.compile(
-				r"[Vv]ulnerability [Dd]escription:(.*?)(?:\n\n[Ii]mpact:|$)",
-				re.DOTALL
-			)
-			impact_pattern = re.compile(
-				r"[Ii]mpact:(.*?)(?:\n\n[Rr]emediation:|$)",
-				re.DOTALL
-			)
-			remediation_pattern = re.compile(
-				r"[Rr]emediation:(.*?)(?:\n\n[Rr]eferences:|$)",
-				re.DOTALL
-			)
+		description_section = extract_between(response_content, vuln_description_pattern)
+		impact_section = extract_between(response_content, impact_pattern)
+		remediation_section = extract_between(response_content, remediation_pattern)
+		references_start_index = response_content.find("References:")
+		references_section = response_content[references_start_index + len("References:"):].strip()
 
-			description_section = extract_between(response_content, vuln_description_pattern)
-			impact_section = extract_between(response_content, impact_pattern)
-			remediation_section = extract_between(response_content, remediation_pattern)
-			references_start_index = response_content.find("References:")
-			references_section = response_content[references_start_index + len("References:"):].strip()
-
-			url_pattern = re.compile(r'https://\S+')
-			urls = url_pattern.findall(references_section)
-
-			return {
-				'status': True,
-				'description': description_section,
-				'impact': impact_section,
-				'remediation': remediation_section,
-				'references': urls,
-			}
-		except Exception as e:
-			return {
-				'status': False,
-				'error': str(e)
-			}
+		url_pattern = re.compile(r'https://\S+')
+		urls = url_pattern.findall(references_section)
 
+		return {
+			'status': True,
+			'description': description_section,
+			'impact': impact_section,
+			'remediation': remediation_section,
+			'references': urls,
+		}
 
 class GPTAttackSuggestionGenerator:
 
 	def __init__(self):
 		self.api_key = get_open_ai_key()
 		self.model_name = 'gpt-3.5-turbo'
+		if not self.api_key:
+			self.ollama = Ollama(base_url='http://ollama:11434', model="llama2-uncensored")
 
 	def get_attack_suggestion(self, input):
 		'''
 			input (str): input for gpt
 		'''
 		if not self.api_key:
-			return {
-				'status': False,
-				'error': 'No OpenAI keys provided.',
-				'input': input
-			}
-		openai.api_key = self.api_key
-		print(input)
-		try:
-			gpt_response = openai.ChatCompletion.create(
-			model=self.model_name,
-			messages=[
-					{'role': 'system', 'content': ATTACK_SUGGESTION_GPT_SYSTEM_PROMPT},
-					{'role': 'user', 'content': input}
-				]
-			)
-			response_content = gpt_response['choices'][0]['message']['content']
-			return {
-				'status': True,
-				'description': response_content,
-				'input': input
-			}
-		except Exception as e:
-			return {
-				'status': False,
-				'error': str(e),
-				'input': input
-			}
+			prompt = ATTACK_SUGGESTION_GPT_SYSTEM_PROMPT + "\nUser: " + input
+			response_content = self.ollama(prompt)
+		else:
+			openai.api_key = self.api_key
+			print(input)
+			try:
+				gpt_response = openai.ChatCompletion.create(
+				model=self.model_name,
+				messages=[
+						{'role': 'system', 'content': ATTACK_SUGGESTION_GPT_SYSTEM_PROMPT},
+						{'role': 'user', 'content': input}
+					]
+				)
+				response_content = gpt_response['choices'][0]['message']['content']
+			except Exception as e:
+				return {
+					'status': False,
+					'error': str(e),
+					'input': input
+				}
+		return {
+			'status': True,
+			'description': response_content,
+			'input': input
+		}
+		
\ No newline at end of file
diff --git a/web/requirements.txt b/web/requirements.txt
index 49966c967..1d5eb5faa 100644
--- a/web/requirements.txt
+++ b/web/requirements.txt
@@ -37,3 +37,4 @@ whatportis
 weasyprint==53.3
 wafw00f==2.2.0
 xmltodict==0.13.0
+langchain==0.0.343
\ No newline at end of file