Initial commit: GenAI Stack with Gemini integration

.gitignore

@@ -1,5 +1,48 @@
+# Environment variables
 .env
+.env.*
+
+# Python
+__pycache__/
+*.py[cod]
+*$py.class
+*.so
+.Python
+env/
+build/
+develop-eggs/
+dist/
+downloads/
+eggs/
+.eggs/
+lib/
+lib64/
+parts/
+sdist/
+var/
+wheels/
+*.egg-info/
+.installed.cfg
+*.egg
+
+# Node
+node_modules/
+npm-debug.log*
+yarn-debug.log*
+yarn-error.log*
+
+# IDE
+.idea/
+.vscode/
+*.swp
+*.swo
+
+# OS
+.DS_Store
+Thumbs.db
+
+# Project specific
 data/
 embedding_model/*
 !embedding_model/.ignore
-.DS_Store
+*.log

chains.py

@@ -6,6 +6,7 @@
 from langchain_openai import ChatOpenAI
 from langchain_ollama import ChatOllama
 from langchain_aws import ChatBedrock
+from langchain_google_genai import ChatGoogleGenerativeAI
 
 from langchain_neo4j import Neo4jVector
 
@@ -81,18 +82,24 @@ def load_llm(llm_name: str, logger=BaseLogger(), config={}):
             model_kwargs={"temperature": 0.0, "max_tokens_to_sample": 1024},
             streaming=True,
         )
-
+    elif llm_name.startswith("gemini"):
+        logger.info(f"LLM: Using Gemini: {llm_name}")
+        return ChatGoogleGenerativeAI(
+            model=llm_name,
+            temperature=0,
+            streaming=True,
+            convert_system_message_to_human=True
+        )
     elif len(llm_name):
         logger.info(f"LLM: Using Ollama: {llm_name}")
         return ChatOllama(
             temperature=0,
             base_url=config["ollama_base_url"],
             model=llm_name,
             streaming=True,
-            # seed=2,
-            top_k=10,  # A higher value (100) will give more diverse answers, while a lower value (10) will be more conservative.
-            top_p=0.3,  # Higher value (0.95) will lead to more diverse text, while a lower value (0.5) will generate more focused text.
-            num_ctx=3072,  # Sets the size of the context window used to generate the next token.
+            top_k=10,
+            top_p=0.3,
+            num_ctx=3072,
         )
     logger.info("LLM: Using GPT-3.5")
     return ChatOpenAI(temperature=0, model_name="gpt-3.5-turbo", streaming=True)

docker-compose.yml

@@ -40,7 +40,7 @@ services:
     environment:
       - NEO4J_AUTH=${NEO4J_USERNAME-neo4j}/${NEO4J_PASSWORD-password}
       - NEO4J_PLUGINS=["apoc"]
-      - NEO4J_db_tx__log_rotation_retention__policy=false
+      - NEO4J_db_tx__log_rotation__retention__policy=false
       - NEO4J_dbms_security_procedures_unrestricted=apoc.*
     healthcheck:
         test: ["CMD-SHELL", "wget --no-verbose --tries=1 --spider localhost:7474 || exit 1"]
@@ -62,8 +62,7 @@ services:
       - NEO4J_USERNAME=${NEO4J_USERNAME-neo4j}
       - OPENAI_API_KEY=${OPENAI_API_KEY-}
       - GOOGLE_API_KEY=${GOOGLE_API_KEY-}      
-      - OLLAMA_BASE_URL=${OLLAMA_BASE_URL-http://host.docker.internal:11434}
-      - EMBEDDING_MODEL=${EMBEDDING_MODEL-sentence_transformer}
+      - EMBEDDING_MODEL=${EMBEDDING_MODEL-google-genai-embedding-001}
       - LANGCHAIN_ENDPOINT=${LANGCHAIN_ENDPOINT-"https://api.smith.langchain.com"}
       - LANGCHAIN_TRACING_V2=${LANGCHAIN_TRACING_V2-false}
       - LANGCHAIN_PROJECT=${LANGCHAIN_PROJECT}
@@ -76,8 +75,6 @@ services:
     depends_on:
       database:
         condition: service_healthy
-      pull-model:
-        condition: service_completed_successfully
     x-develop:
       watch:
         - action: rebuild
@@ -104,9 +101,8 @@ services:
       - NEO4J_USERNAME=${NEO4J_USERNAME-neo4j}
       - OPENAI_API_KEY=${OPENAI_API_KEY-}      
       - GOOGLE_API_KEY=${GOOGLE_API_KEY-}
-      - OLLAMA_BASE_URL=${OLLAMA_BASE_URL-http://host.docker.internal:11434}
-      - LLM=${LLM-llama2}
-      - EMBEDDING_MODEL=${EMBEDDING_MODEL-sentence_transformer}
+      - LLM=${LLM-gemini-pro}
+      - EMBEDDING_MODEL=${EMBEDDING_MODEL-google-genai-embedding-001}
       - LANGCHAIN_ENDPOINT=${LANGCHAIN_ENDPOINT-"https://api.smith.langchain.com"}
       - LANGCHAIN_TRACING_V2=${LANGCHAIN_TRACING_V2-false}
       - LANGCHAIN_PROJECT=${LANGCHAIN_PROJECT}
@@ -119,8 +115,6 @@ services:
     depends_on:
       database:
         condition: service_healthy
-      pull-model:
-        condition: service_completed_successfully
     x-develop:
       watch:
         - action: rebuild
@@ -143,9 +137,8 @@ services:
       - NEO4J_USERNAME=${NEO4J_USERNAME-neo4j}
       - OPENAI_API_KEY=${OPENAI_API_KEY-}
       - GOOGLE_API_KEY=${GOOGLE_API_KEY-}
-      - OLLAMA_BASE_URL=${OLLAMA_BASE_URL-http://host.docker.internal:11434}
-      - LLM=${LLM-llama2}
-      - EMBEDDING_MODEL=${EMBEDDING_MODEL-sentence_transformer}
+      - LLM=${LLM-gemini-pro}
+      - EMBEDDING_MODEL=${EMBEDDING_MODEL-google-genai-embedding-001}
       - LANGCHAIN_ENDPOINT=${LANGCHAIN_ENDPOINT-"https://api.smith.langchain.com"}
       - LANGCHAIN_TRACING_V2=${LANGCHAIN_TRACING_V2-false}
       - LANGCHAIN_PROJECT=${LANGCHAIN_PROJECT}
@@ -158,8 +151,6 @@ services:
     depends_on:
       database:
         condition: service_healthy
-      pull-model:
-        condition: service_completed_successfully
     x-develop:
       watch:
         - action: rebuild
@@ -184,9 +175,8 @@ services:
       - NEO4J_USERNAME=${NEO4J_USERNAME-neo4j}
       - OPENAI_API_KEY=${OPENAI_API_KEY}
       - GOOGLE_API_KEY=${GOOGLE_API_KEY}  
-      - OLLAMA_BASE_URL=${OLLAMA_BASE_URL-http://host.docker.internal:11434}
-      - LLM=${LLM-llama2}
-      - EMBEDDING_MODEL=${EMBEDDING_MODEL-sentence_transformer}
+      - LLM=${LLM-gemini-pro}
+      - EMBEDDING_MODEL=${EMBEDDING_MODEL-google-genai-embedding-001}
       - LANGCHAIN_ENDPOINT=${LANGCHAIN_ENDPOINT-"https://api.smith.langchain.com"}
       - LANGCHAIN_TRACING_V2=${LANGCHAIN_TRACING_V2-false}
       - LANGCHAIN_PROJECT=${LANGCHAIN_PROJECT}
@@ -199,8 +189,6 @@ services:
     depends_on:
       database:
         condition: service_healthy
-      pull-model:
-        condition: service_completed_successfully
     x-develop:
       watch:
         - action: rebuild

front-end/app.py

@@ -0,0 +1,189 @@
+from flask import Flask, render_template, request, jsonify
+import os
+from dotenv import load_dotenv
+import openai
+import google.generativeai as genai
+import anthropic
+from langchain_community.llms import Ollama
+
+app = Flask(__name__)
+load_dotenv()
+
+# Model configurations
+MODEL_CONFIGS = {
+    'openai': {
+        'name': 'OpenAI',
+        'models': [
+            'gpt-4-turbo-preview',
+            'gpt-4',
+            'gpt-3.5-turbo',
+            'gpt-3.5-turbo-16k'
+        ],
+        'api_key_env': 'OPENAI_API_KEY'
+    },
+    'gemini': {
+        'name': 'Google Gemini',
+        'models': [
+            'gemini-pro',
+            'gemini-pro-vision',
+            'gemini-2.0-flash'
+
+        ],
+        'api_key_env': 'GOOGLE_API_KEY'
+    },
+    'claude': {
+        'name': 'Anthropic Claude',
+        'models': [
+            'claude-3-opus-20240229',
+            'claude-3-sonnet-20240229',
+            'claude-3-haiku-20240307'
+        ],
+        'api_key_env': 'ANTHROPIC_API_KEY'
+    },
+    'ollama': {
+        'name': 'Ollama',
+        'models': [
+            'llama2',
+            'llama2:13b',
+            'llama2:70b',
+            'mistral',
+            'mistral-openorca',
+            'codellama',
+            'codellama:13b',
+            'codellama:34b',
+            'neural-chat',
+            'starling-lm',
+            'dolphin-phi',
+            'orca-mini',
+            'vicuna',
+            'wizard-vicuna-uncensored'
+        ],
+        'api_key_env': None
+    },
+    'cohere': {
+        'name': 'Cohere',
+        'models': [
+            'command',
+            'command-light',
+            'command-r',
+            'command-r-plus'
+        ],
+        'api_key_env': 'COHERE_API_KEY'
+    },
+    'huggingface': {
+        'name': 'Hugging Face',
+        'models': [
+            'meta-llama/Llama-2-7b-chat-hf',
+            'meta-llama/Llama-2-13b-chat-hf',
+            'meta-llama/Llama-2-70b-chat-hf',
+            'mistralai/Mistral-7B-Instruct-v0.2',
+            'google/flan-t5-xxl',
+            'google/flan-ul2'
+        ],
+        'api_key_env': 'HUGGINGFACE_API_KEY'
+    }
+}
+
+def get_available_models():
+    available_models = []
+    for provider, config in MODEL_CONFIGS.items():
+        if config['api_key_env'] is None or os.getenv(config['api_key_env']):
+            available_models.append({
+                'provider': provider,
+                'name': config['name'],
+                'models': config['models']
+            })
+    return available_models
+
+@app.route('/')
+def index():
+    return render_template('index.html')
+
+@app.route('/api/models')
+def get_models():
+    return jsonify(get_available_models())
+
+@app.route('/api/chat', methods=['POST'])
+def chat():
+    data = request.json
+    message = data.get('message')
+    model = data.get('model')
+    provider = data.get('provider')
+    temperature = float(data.get('temperature', 0.7))
+    max_tokens = int(data.get('max_tokens', 2000))
+
+    if not message or not model or not provider:
+        return jsonify({'error': 'Missing required parameters'}), 400
+
+    try:
+        if provider == 'openai':
+            client = openai.OpenAI(api_key=os.getenv('OPENAI_API_KEY'))
+            response = client.chat.completions.create(
+                model=model,
+                messages=[{"role": "user", "content": message}],
+                temperature=temperature,
+                max_tokens=max_tokens
+            )
+            return jsonify({'response': response.choices[0].message.content})
+
+        elif provider == 'gemini':
+            genai.configure(api_key=os.getenv('GOOGLE_API_KEY'))
+            model = genai.GenerativeModel(model)
+            response = model.generate_content(
+                message,
+                generation_config=genai.types.GenerationConfig(
+                    temperature=temperature,
+                    max_output_tokens=max_tokens
+                )
+            )
+            return jsonify({'response': response.text})
+
+        elif provider == 'claude':
+            client = anthropic.Anthropic(api_key=os.getenv('ANTHROPIC_API_KEY'))
+            response = client.messages.create(
+                model=model,
+                max_tokens=max_tokens,
+                temperature=temperature,
+                messages=[{"role": "user", "content": message}]
+            )
+            return jsonify({'response': response.content[0].text})
+
+        elif provider == 'ollama':
+            llm = Ollama(
+                model=model,
+                temperature=temperature,
+                num_predict=max_tokens
+            )
+            response = llm.invoke(message)
+            return jsonify({'response': response})
+
+        elif provider == 'cohere':
+            import cohere
+            co = cohere.Client(os.getenv('COHERE_API_KEY'))
+            response = co.generate(
+                prompt=message,
+                model=model,
+                temperature=temperature,
+                max_tokens=max_tokens
+            )
+            return jsonify({'response': response.generations[0].text})
+
+        elif provider == 'huggingface':
+            from huggingface_hub import InferenceClient
+            client = InferenceClient(token=os.getenv('HUGGINGFACE_API_KEY'))
+            response = client.text_generation(
+                message,
+                model=model,
+                max_new_tokens=max_tokens,
+                temperature=temperature
+            )
+            return jsonify({'response': response})
+
+        else:
+            return jsonify({'error': 'Unsupported provider'}), 400
+
+    except Exception as e:
+        return jsonify({'error': str(e)}), 500
+
+if __name__ == '__main__':
+    app.run(debug=True, port=5001)