ollama

module  pixeltable.functions.ollama

Pixeltable UDFs for Ollama local models.

Provides integration with Ollama for running large language models locally, including chat completions and embeddings.

udf  chat()

Signature

@pxt.udf
chat(messages: pxt.Json[(Json, *, model: pxt.String, tools: pxt.Json[(Json = None, format: pxt.String | None = None, options: pxt.Json | None = None) -> pxt.Json

Generate the next message in a chat with a provided model.

Parameters:

• messages (pxt.Json[(Json): The messages of the chat.
• model (Any): The model name.
• tools (Any): Tools for the model to use.
• format (Any): The format of the response; must be one of 'json' or None.
• options (Any): Additional options to pass to the chat call, such as max_tokens, temperature, top_p, and top_k. For details, see the Valid Parameters and Values section of the Ollama documentation.

udf  embed()

Signature

@pxt.udf
embed(
    input: pxt.String,
    *,
    model: pxt.String,
    truncate: pxt.Bool = True,
    options: pxt.Json | None = None
) -> pxt.Array[(None,), float32]

Generate embeddings from a model.

Parameters:

• input (pxt.String): The input text to generate embeddings for.
• model (pxt.String): The model name.
• truncate (pxt.Bool): Truncates the end of each input to fit within context length. Returns error if false and context length is exceeded.
• options (pxt.Json | None): Additional options to pass to the embed call. For details, see the Valid Parameters and Values section of the Ollama documentation.

udf  generate()

Signature

@pxt.udf
generate(
    prompt: pxt.String,
    *,
    model: pxt.String,
    suffix: pxt.String = '',
    system: pxt.String = '',
    template: pxt.String = '',
    context: pxt.Json[(Int = None, raw: pxt.Bool = False, format: pxt.String | None = None, options: pxt.Json | None = None
) -> pxt.Json

Generate a response for a given prompt with a provided model.

Parameters:

• prompt (pxt.String): The prompt to generate a response for.
• model (pxt.String): The model name.
• suffix (pxt.String): The text after the model response.
• format (Any): The format of the response; must be one of 'json' or None.
• system (pxt.String): System message.
• template (pxt.String): Prompt template to use.
• context (pxt.Json[(Int): The context parameter returned from a previous call to generate().
• raw (Any): If True, no formatting will be applied to the prompt.
• options (Any): Additional options for the Ollama chat call, such as max_tokens, temperature, top_p, and top_k. For details, see the Valid Parameters and Values section of the Ollama documentation.