Step-by-step guide for integrating new language models into the Python, Node.js, and Next.js stacks.

​
Integration Types

​
OpenRouter Integration

For models accessed through OpenRouter (RAG, Simple Tool Chat, CustomGPT)

​
Direct SDK Integration

For direct API integration (OpenAI, Claude, Groq, Mistral)

​
Python Stack Integration

​
OpenRouter-Based Models

Service Layer: src/chatflow_langchain/service/
  • Add model service folders (custom_gpt, title, simple_chat)
Controller Logic: src/chatflow_langchain/controller/
  • Update controllers for OpenRouter registration and routing
Prompt Templates: src/prompt/langchain/
  • Define prompt templates and configuration
Callbacks: src/customlib/langchain/
  • Implement MongoDB persistence
  • Token usage tracking
  • Streaming response management

​
Direct SDK Integration

Dependencies: requirements.txt
  • Add required SDKs (openai, anthropic, mistral)
Model Configuration: src/chatflow_langchain/service/model/config.py
  • Add default behavior and token limits
  • Define model metadata
Optional Caching: chatopenai_cache.py
  • Add caching logic (OpenAI-specific)
  • Usage tracking and refresh/expiry logic
Controller Updates: src/chatflow_langchain/controller/
  • Add routing and provider logic
Prompt Updates: src/prompt/langchain/
  • Align templates for model behavior

​
Node.js Backend Configuration

​
Constants Configuration

AI Model Constants: src/config/constants/aimodal.js
  • Add new model constant
  • Update OPENROUTER_PROVIDER, MODAL_NAME
Common Constants: src/config/constants/common.js
  • Add model code in MODEL_CODE

​
API Key Management

Company Service: src/services/company.js
  • Extend checkApiKey() for new model support
  • Define API validation checker function

​
Next.js Frontend Integration

​
Assets and Constants

Model Icon: Place logo in public/ directory Constant Mapping: src/utils/constant.ts
  • Add model code, display name, image path
  • Configure credit info and sequence list

​
Feature Support

Helper Functions: src/utils/helper.ts
  • Update allowImageGeneration if supported
  • Update allowImageConversation if supported

​
Database Schema

​
Model Collection

Global model metadata: 
{
  "title": "Model Name",
  "code": "MODEL_CODE", 
  "isActive": true
}

​
CompanyModel Collection

Company-specific model configuration: 
{
  "name": "model-name",
  "bot": {
    "title": "Model Name",
    "code": "MODEL_CODE",
    "id": "ModelCollectionId"
  },
  "company": {
    "name": "Company Name",
    "slug": "company-slug", 
    "id": "CompanyId"
  },
  "config": {
    "apikey": "your-api-key"
  },
  "modelType": 2,
  "dimensions": 1536,
  "isActive": true,
  "extraConfig": {
    "stop": [],
    "temperature": 0.7,
    "tools": []
  }
}

​
Integration Checklist

ComponentPurposePath
Service LayerModel-specific servicessrc/chatflow_langchain/service/
ControllersModel routing and registrationsrc/chatflow_langchain/controller/
Prompt TemplatesModel prompt configurationsrc/prompt/langchain/
CallbacksToken tracking and persistencesrc/customlib/langchain/
DependenciesRequired SDKsrequirements.txt
Node ConstantsModel constants and codessrc/config/constants/
API Key LogicValidation and checkingsrc/services/company.js
Frontend ConstantsUI metadata and configurationsrc/utils/constant.ts
Model AssetsLogos and iconspublic/

​
Testing Requirements

  • End-to-end testing from model selection to response
  • Prompt behavior validation
  • Tool compatibility verification
  • API key validation testing
  • Cost tracking accuracy
Always test the complete integration flow including model selection, prompt processing, response generation, and cost tracking before deployment.