| Availability |
Odoo Online
Odoo.sh
On Premise
|
| Odoo Apps Dependencies |
Discuss (mail)
|
| Community Apps Dependencies | Show |
| Lines of code | 4372 |
| Technical Name |
llm_openai |
| License | LGPL-3 |
| Website | https://github.com/apexive/odoo-llm |
| Versions | 16.0 18.0 |
| Availability |
Odoo Online
Odoo.sh
On Premise
|
| Odoo Apps Dependencies |
Discuss (mail)
|
| Community Apps Dependencies | Show |
| Lines of code | 4372 |
| Technical Name |
llm_openai |
| License | LGPL-3 |
| Website | https://github.com/apexive/odoo-llm |
| Versions | 16.0 18.0 |
OpenAI Provider for Odoo LLM
Connect your Odoo instance with OpenAI's powerful models like GPT-4, GPT-3.5, and more.
This module provides a seamless integration with OpenAI API and supports OpenAI-compatible endpoints from various providers.
Overview
The OpenAI Provider module extends the LLM Integration Base to connect with OpenAI's powerful models. This module allows you to use ChatGPT, GPT-4, and other OpenAI models directly from your Odoo instance.
Key Capabilities
- Chat completions - Generate conversational responses using GPT models
- Text embeddings - Create vector representations of text using OpenAI's embedding models
- Function calling - Enable AI models to execute functions through a standardized interface
- Multi-provider support - Connect to OpenAI-compatible endpoints from various providers
Configuration
Setting up the OpenAI provider is straightforward:
- Install the module in your Odoo instance
- Navigate to LLM > Configuration > Providers
- Create a new provider and select "OpenAI" as the provider type
- Enter your OpenAI API key
- Click "Fetch Models" to import available models
- Set default models for chat, embedding, and other services
OpenAI-Compatible Endpoints
Many AI providers offer OpenAI-compatible endpoints, allowing you to use the same code and configuration with different services. This module supports connecting to any of these compatible endpoints by changing the base URL in the provider configuration.
Supported OpenAI-Compatible Endpoints
| Provider | Base URL | Models | Notes |
|---|---|---|---|
| OpenAI | https://api.openai.com/v1/ |
GPT-4o, GPT-4, GPT-3.5 Turbo, etc. | Official OpenAI API |
| Anthropic | https://api.anthropic.com/v1/ |
Claude 3 Opus, Sonnet, Haiku | Requires Anthropic API key |
| DeepSeek | https://api.deepseek.com/v1/ |
DeepSeek-Coder, DeepSeek-Chat | Requires DeepSeek API key |
| Ollama | http://localhost:11434/v1/ |
Llama, Mistral, Vicuna, etc. | Local deployment, no API key needed |
| Google AI | https://generativelanguage.googleapis.com/v1beta/openai/ |
Gemini models | Requires Google API key |
| Mistral AI | https://api.mistral.ai/v1/ |
Mistral-7B, Mixtral-8x7B, etc. | Requires Mistral AI API key |
Using OpenAI-Compatible Endpoints
To use an alternative provider with an OpenAI-compatible endpoint:
- Create a new provider in LLM > Configuration > Providers
- Select "OpenAI" as the provider type
- Enter your API key for the specific provider
- Set the "Base URL" field to the appropriate endpoint URL from the table above
- Click "Fetch Models" to discover available models
Provider Compatibility Note
While these providers offer OpenAI-compatible endpoints, there may be slight differences in API behavior or supported features. Always refer to the provider's documentation for specific details about their OpenAI compatibility.
Technical Details
This module extends the base LLM integration framework with OpenAI-specific implementations:
- Implements the OpenAI API client with proper authentication and error handling
- Provides model mapping and conversion between OpenAI formats and Odoo LLM formats
- Handles rate limiting and retries according to OpenAI best practices
- Supports streaming responses for chat completions
Module Information
| Name | OpenAI Provider for Odoo LLM |
|---|---|
| Version | 18.0.1.1.2 |
| Category | Technical |
| License | LGPL-3 |
| Dependencies | llm, llm_tool, llm_mail_message_subtypes |
| External Dependencies | openai (Python package) |
| Author | Apexive Solutions LLC |
Please log in to comment on this module