Class OptionsBuilder
java.lang.Object
io.github.amithkoujalgi.ollama4j.core.utils.OptionsBuilder
Builder class for creating options for Ollama model.
-
Constructor Summary
Constructors -
Method Summary
Modifier and TypeMethodDescriptionbuild()
Builds the options map.setMirostat
(int value) Enable Mirostat sampling for controlling perplexity.setMirostatEta
(float value) Influences how quickly the algorithm responds to feedback from the generated text.setMirostatTau
(float value) Controls the balance between coherence and diversity of the output.setNumCtx
(int value) Sets the size of the context window used to generate the next token.setNumGpu
(int value) The number of layers to send to the GPU(s).setNumGqa
(int value) The number of GQA groups in the transformer layer.setNumPredict
(int value) Maximum number of tokens to predict when generating text.setNumThread
(int value) Sets the number of threads to use during computation.setRepeatLastN
(int value) Sets how far back for the model to look back to prevent repetition.setRepeatPenalty
(float value) Sets how strongly to penalize repetitions.setSeed
(int value) Sets the random number seed to use for generation.Sets the stop sequences to use.setTemperature
(float value) The temperature of the model.setTfsZ
(float value) Tail free sampling is used to reduce the impact of less probable tokens from the output.setTopK
(int value) Reduces the probability of generating nonsense.setTopP
(float value) Works together with top-k.
-
Constructor Details
-
OptionsBuilder
public OptionsBuilder()Constructs a new OptionsBuilder with an empty options map.
-
-
Method Details
-
setMirostat
Enable Mirostat sampling for controlling perplexity. (default: 0, 0 = disabled, 1 = Mirostat, 2 = Mirostat 2.0)- Parameters:
value
- The value for the "mirostat" parameter.- Returns:
- The updated OptionsBuilder.
-
setMirostatEta
Influences how quickly the algorithm responds to feedback from the generated text. A lower learning rate will result in slower adjustments, while a higher learning rate will make the algorithm more responsive. (Default: 0.1)- Parameters:
value
- The value for the "mirostat_eta" parameter.- Returns:
- The updated OptionsBuilder.
-
setMirostatTau
Controls the balance between coherence and diversity of the output. A lower value will result in more focused and coherent text. (Default: 5.0)- Parameters:
value
- The value for the "mirostat_tau" parameter.- Returns:
- The updated OptionsBuilder.
-
setNumCtx
Sets the size of the context window used to generate the next token. (Default: 2048)- Parameters:
value
- The value for the "num_ctx" parameter.- Returns:
- The updated OptionsBuilder.
-
setNumGqa
The number of GQA groups in the transformer layer. Required for some models, for example, it is 8 for llama2:70b.- Parameters:
value
- The value for the "num_gqa" parameter.- Returns:
- The updated OptionsBuilder.
-
setNumGpu
The number of layers to send to the GPU(s). On macOS it defaults to 1 to enable metal support, 0 to disable.- Parameters:
value
- The value for the "num_gpu" parameter.- Returns:
- The updated OptionsBuilder.
-
setNumThread
Sets the number of threads to use during computation. By default, Ollama will detect this for optimal performance. It is recommended to set this value to the number of physical CPU cores your system has (as opposed to the logical number of cores).- Parameters:
value
- The value for the "num_thread" parameter.- Returns:
- The updated OptionsBuilder.
-
setRepeatLastN
Sets how far back for the model to look back to prevent repetition. (Default: 64, 0 = disabled, -1 = num_ctx)- Parameters:
value
- The value for the "repeat_last_n" parameter.- Returns:
- The updated OptionsBuilder.
-
setRepeatPenalty
Sets how strongly to penalize repetitions. A higher value (e.g., 1.5) will penalize repetitions more strongly, while a lower value (e.g., 0.9) will be more lenient. (Default: 1.1)- Parameters:
value
- The value for the "repeat_penalty" parameter.- Returns:
- The updated OptionsBuilder.
-
setTemperature
The temperature of the model. Increasing the temperature will make the model answer more creatively. (Default: 0.8)- Parameters:
value
- The value for the "temperature" parameter.- Returns:
- The updated OptionsBuilder.
-
setSeed
Sets the random number seed to use for generation. Setting this to a specific number will make the model generate the same text for the same prompt. (Default: 0)- Parameters:
value
- The value for the "seed" parameter.- Returns:
- The updated OptionsBuilder.
-
setStop
Sets the stop sequences to use. When this pattern is encountered the LLM will stop generating text and return. Multiple stop patterns may be set by specifying multiple separate `stop` parameters in a modelfile.- Parameters:
value
- The value for the "stop" parameter.- Returns:
- The updated OptionsBuilder.
-
setTfsZ
Tail free sampling is used to reduce the impact of less probable tokens from the output. A higher value (e.g., 2.0) will reduce the impact more, while a value of 1.0 disables this setting. (default: 1)- Parameters:
value
- The value for the "tfs_z" parameter.- Returns:
- The updated OptionsBuilder.
-
setNumPredict
Maximum number of tokens to predict when generating text. (Default: 128, -1 = infinite generation, -2 = fill context)- Parameters:
value
- The value for the "num_predict" parameter.- Returns:
- The updated OptionsBuilder.
-
setTopK
Reduces the probability of generating nonsense. A higher value (e.g. 100) will give more diverse answers, while a lower value (e.g. 10) will be more conservative. (Default: 40)- Parameters:
value
- The value for the "top_k" parameter.- Returns:
- The updated OptionsBuilder.
-
setTopP
Works together with top-k. A higher value (e.g., 0.95) will lead to more diverse text, while a lower value (e.g., 0.5) will generate more focused and conservative text. (Default: 0.9)- Parameters:
value
- The value for the "top_p" parameter.- Returns:
- The updated OptionsBuilder.
-
build
Builds the options map.- Returns:
- The populated options map.
-