Parameters
| Parameter | Type | Default | Description |
|---|---|---|---|
base_url | str | "http://localhost:11434" | Ollama server URL |
model_name | str | "nomic-embed-text" | Ollama embedding model name |
auto_pull_model | bool | True | Automatically pull model if not available |
keep_alive | Optional[str] | "5m" | Keep model loaded for duration |
temperature | Optional[float] | None | Model temperature |
top_p | Optional[float] | None | Top-p sampling |
num_ctx | Optional[int] | None | Context window size |
request_timeout | float | 120.0 | Request timeout in seconds |
connection_timeout | float | 10.0 | Connection timeout in seconds |
max_retries | int | 3 | Maximum number of retries |
enable_keep_alive | bool | True | Keep model loaded between requests |
enable_model_preload | bool | True | Preload model on startup |
Functions
__init__
Initialize the OllamaEmbedding provider.
Parameters:
config(Optional[OllamaEmbeddingConfig]): Configuration object**kwargs: Additional configuration options
_setup_http_session
Setup HTTP session for requests.
_initialize_sync
Initialize Ollama connection and model synchronously.
_initialize_async
Initialize Ollama connection and model asynchronously.
_check_ollama_health_sync
Check if Ollama server is healthy synchronously.
Returns:
bool: True if server is healthy
_check_ollama_health
Check if Ollama server is healthy asynchronously.
Returns:
bool: True if server is healthy
_ensure_model_available_sync
Ensure the embedding model is available synchronously.
_ensure_model_available
Ensure the embedding model is available asynchronously.
_list_models_sync
List available models in Ollama synchronously.
Returns:
List[Dict[str, Any]]: List of available models
_list_models
List available models in Ollama asynchronously.
Returns:
List[Dict[str, Any]]: List of available models
_pull_model_sync
Pull the embedding model synchronously.
_pull_model
Pull the embedding model asynchronously.
_preload_model
Preload the model to keep it in memory.
supported_modes
Get supported embedding modes.
Returns:
List[EmbeddingMode]: List of supported embedding modes
pricing_info
Get Ollama pricing info (local execution is free).
Returns:
Dict[str, float]: Pricing information
get_model_info
Get information about the current Ollama model.
Returns:
Dict[str, Any]: Model information
_make_embedding_request
Make embedding request to Ollama API.
Parameters:
texts(List[str]): List of texts to embed
Dict[str, Any]: API response
_embed_batch
Embed a batch of texts using Ollama.
Parameters:
texts(List[str]): List of text strings to embedmode(EmbeddingMode): Embedding mode
List[List[float]]: List of embedding vectors
validate_connection
Validate Ollama connection and model access.
Returns:
bool: True if connection is valid
get_server_info
Get Ollama server information.
Returns:
Dict[str, Any]: Server information
get_model_details
Get detailed information about the current model.
Returns:
Dict[str, Any]: Model details
close
Clean up Ollama HTTP sessions and connections.
__del__
Destructor to ensure session is closed when object is garbage collected.
create_nomic_embedding
Create Nomic Embed Text embedding provider.
Parameters:
**kwargs: Additional configuration options
OllamaEmbedding: Configured OllamaEmbedding instance
create_mxbai_embedding
Create MXBAI Large embedding provider.
Parameters:
**kwargs: Additional configuration options
OllamaEmbedding: Configured OllamaEmbedding instance
create_arctic_embedding
Create Snowflake Arctic embedding provider.
Parameters:
**kwargs: Additional configuration options
OllamaEmbedding: Configured OllamaEmbedding instance
create_ollama_embedding_with_auto_setup
Create Ollama embedding with automatic model setup.
Parameters:
model_name(str): Model name to usebase_url(str): Ollama server URL**kwargs: Additional configuration options
OllamaEmbedding: Configured OllamaEmbedding instance