Module io.helidon.integrations.langchain4j.providers.ollama
Package io.helidon.integrations.langchain4j.providers.ollama

Interface OllamaChatModelConfig

All Superinterfaces:
Prototype.Api
All Known Implementing Classes:
OllamaChatModelConfig.BuilderBase.OllamaChatModelConfigImpl
public interface OllamaChatModelConfig extends Prototype.Api
Configuration class for the Ollama chat model, OllamaChatModel.
See Also:

  • Field Details

  • Method Details

    • builder

      static OllamaChatModelConfig.Builder builder()
      Create a new fluent API builder to customize configuration.
      Returns:
      a new builder

    • builder

      static OllamaChatModelConfig.Builder builder(OllamaChatModelConfig instance)
      Create a new fluent API builder from an existing instance.
      Parameters:
      instance - an existing instance used as a base for the builder
      Returns:
      a builder based on an instance

    • create

      static OllamaChatModelConfig create(Config config)
      Create a new instance from configuration.
      Parameters:
      config - used to configure the new instance
      Returns:
      a new instance configured from configuration

    • create

      static OllamaChatModelConfig create()
      Create a new instance with default values.
      Returns:
      a new instance

    • temperature

      Optional<Double> temperature()
      The sampling temperature to use, between 0 and 2. Higher values make the output more random, while lower values make it more focused and deterministic.
      Returns:
      an Optional containing the sampling temperature

    • topK

      Optional<Integer> topK()
      The maximum number of top-probability tokens to consider when generating text. Limits the token pool to the topK highest-probability tokens, controlling the balance between deterministic and diverse outputs.

      A smaller topK (e.g., 1) results in deterministic output, while a larger value (e.g., 50) allows for more variability and creativity.

      Returns:
      an Optional containing the topK value

    • topP

      Optional<Double> topP()
      The nucleus sampling value, where the model considers the results of the tokens with top_p probability mass.
      Returns:
      an Optional containing the nucleus sampling value

    • repeatPenalty

      Optional<Double> repeatPenalty()
      The penalty applied to repeated tokens during text generation. Higher values discourage the model from generating the same token multiple times, promoting more varied and natural output.

      A value of 1.0 applies no penalty (default behavior), while values greater than 1.0 reduce the likelihood of repetition. Excessively high values may overly penalize common phrases, leading to unnatural results.

      Returns:
      an Optional containing the repeat penalty

    • seed

      Optional<Integer> seed()
      The seed for the random number generator used by the model.
      Returns:
      an Optional containing the seed

    • numPredict

      Optional<Integer> numPredict()
      The number of tokens to generate during text prediction. This parameter determines the length of the output generated by the model.
      Returns:
      an Optional containing the numPredicts value

    • stop

      List<String> stop()
      The list of sequences where the API will stop generating further tokens.
      Returns:
      the list of stop sequences

    • format

      Optional<String> format()
      The format of the generated output. This parameter specifies the structure or style of the text produced by the model, such as plain text, JSON, or a custom format.

      Common examples:

      • "plain": Generates unstructured plain text.
      • "json": Produces output formatted as a JSON object.
      • Custom values may be supported depending on the model's capabilities.
      Returns:
      an Optional containing the response format

    • enabled

      boolean enabled()
      If set to false (default), Ollama model will not be available even if configured.
      Returns:
      whether Ollama model is enabled, defaults to false

    • baseUrl

      Optional<String> baseUrl()
      The base URL for the Ollama API.
      Returns:
      the base URL

    • logRequests

      Optional<Boolean> logRequests()
      Whether to log API requests.
      Returns:
      an Optional containing true if requests should be logged, false otherwise

    • logResponses

      Optional<Boolean> logResponses()
      Whether to log API responses.
      Returns:
      an Optional containing true if responses should be logged, false otherwise

    • customHeaders

      Map<String,String> customHeaders()
      A map containing custom headers.
      Returns:
      custom headers map

    • timeout

      Optional<Duration> timeout()
      The timeout setting for API requests.
      Returns:
      the timeout setting in Duration.parse(java.lang.CharSequence) format

    • maxRetries

      Optional<Integer> maxRetries()
      The maximum number of retries for failed API requests.
      Returns:
      an Optional containing the maximum number of retries

    • modelName

      Optional<String> modelName()
      The name of the Ollama model to use. This parameter determines which pre-trained model will process the input prompt and produce the output.

      Examples of valid model names:

      • "llama-2": Utilizes the LLaMA 2 model.
      • "alpaca": Uses a fine-tuned LLaMA model for conversational tasks.
      • "custom-model": A user-defined model trained for specific use cases.
      Returns:
      an Optional containing the model name