Whatever message this page gives is out now! Go check it out!
EmbeddingModel
Last update:
May 18, 2026
Description
Initializes a new Embedding Model client instance. This function acts as the factory for connecting to various embedding providers (such as OpenAI, Azure OpenAI, local models, etc.). It encapsulates model configuration (model name, dimension, provider) and connection settings.
Syntax
EmbeddingModel(struct configuration)Returns
An EmbeddingModel object used to generate embeddings for text input.
Parameters
Parameter | Type | Required | Description |
configuration | Struct | Yes | Structure containing provider and model configuration. |
The configuration structure supports the following common keys. Exact support varies by provider (OpenAI, AzureOpenAI, local, etc.).
Provider parameters
Provider | Required | Optional |
OpenAI | API Key Model Name | Base URL Dimension Organization ID Project ID User Batch Size Max Retries Timeout (ms) Log Requests Log Responses |
Azure OpenAI | API Key Deployment Name | Base URL Dimension Max Retries Timeout (ms) Log Requests Log Responses |
Mistral | API Key Model Name | Base URL Max Retries Timeout (ms) Log Requests Log Responses |
Ollama | Model Name | Base URL Max Retries Timeout (ms) Log Requests Log Responses |
Gemini | API Key Model Name | Base URL Dimension Max Retries Timeout (ms) Log Requests Log Responses |
Key | Type | Default | Description |
provider | String | - | Embedding provider. Examples: "openai", "azureOpenAI", "ollama". |
apiKey | String | - | API key or token for the provider (if required). |
baseUrl | String | - | Optional base endpoint URL (for self‑hosted or Azure endpoints). |
modelName | String | - | Embedding model identifier. Example: "text-embedding-3-small". |
dimension | Numeric | - | Expected embedding dimension (for validation). |
timeout | Numeric | 10000 | Request timeout in milliseconds. |
maxRetries | Numeric | 2 | Maximum retry attempts for transient failures. |
logRequests | Boolean | true | Whether to log outbound requests (for debugging). |
logResponses | Boolean | true | Whether to log responses (for debugging). |
metadata | Struct | {} | Optional key‑value metadata to tag embedding calls. |
Usage
Use EmbeddingModel once during initialization. The returned embeddingModel instance exposes: embed(text) – generate an embedding for a single text string.
- embedAll(textArray) – generate embeddings for an array of text strings.
Example
<cfscript>
embeddingModelConfig = EmbeddingModel(
{
provider : "openAI",
modelName : "text-embedding-3-small",
apiKey : "#application.apiKey#"
}
)
//embeddingModelConfig.embed("The quick brown fox jumps over the lazy dog")
//writeDump(response)
// add a vector store
inMemoryClient=VectorStore({
provider:"inMemory"
});
// add a single item to vector store
response=inMemoryClient.add({
id: createUUID(),
embedding: embeddingModelConfig,
text: "ColdFusion is a powerful rapid application development platform",
metadata: {"category": "technology", "test": "add"}
})
writeDump(response)
</cfscript>
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
|
