Whatever message this page gives is out now! Go check it out!
Introduction to models in ColdFusion
Last update:
May 18, 2026
Introduction
ColdFusion 2025.0.08 introduces a vendor-neutral framework for integrating generative AI into enterprise applications. This framework abstracts provider-specific complexities, enabling developers to build agents, chatbots, and semantic search pipelines using standard CFML.
Switching between models or providers is as simple as updating configuration, with no need to refactor application code.
Why use ColdFusion AI
Treating AI providers as infrastructure offers several advantages:
- Vendor independence: Switch providers by updating configuration.
- Data privacy: Support for local inference servers (e.g., Ollama) keeps sensitive data off the cloud.
- Unified tooling: Standardized features like function calling and retrieval-augmented generation (RAG) work consistently across all providers.
- Orchestration: Built-in support for Model Context Protocol (MCP) for smooth integration with external tools and services.
Core architecture
The framework is layered:
- LLMModel layer: Stateless, direct access to an LLM for text generation, translation, or code assistance. Instantiated using
ChatModel(). a - Agent layer: Stateful wrapper that manages conversation history (memory), personas (system messages), and tool execution. Instantiated using
Agent(). - RAG pipeline: End-to-end solution for document ingestion and retrieval, including document loaders, text splitters, embedding models, and vector stores.
Supported providers
ColdFusion 2025.0.08 supports the following AI providers:
Online
- OpenAI
- Anthropic
- Mistral
- Google Gemini
Cloud-hosted
- Azure OpenAI
Offline
- Ollama
Pre-requisites
- ColdFusion version: ColdFusion 2025 Update 7 (Beta)
- Network: Outbound access to provider APIs (for online and cloud-hosted providers), or a running local inference server (for Ollama).
- Credentials: API keys for online and cloud-hosted providers. Ollama does not require an API key.
- After installing the update, install the ai package using the ColdFusion Package Manager or from the ColdFusion Administrator.
We're glad. Tell us how this page helped.
We're sorry. Can you tell us what didn't work for you?
Share additional feedback to help us improve.
0/255
|
