inference-config

Manage inference configuration settings for text generation. This command allows you to view, modify, and reset generation parameters like temperature, top-p, and max tokens.

Basic Usage
python -m wisent inference-config [show|set|reset] [OPTIONS]

Examples

Show Current Config
python -m wisent inference-config show
Set Temperature
python -m wisent inference-config set --temperature 0.7
Set Multiple Parameters
python -m wisent inference-config set \
  --temperature 0.8 \
  --top-p 0.95 \
  --max-new-tokens 256 \
  --do-sample true
Enable Thinking Mode (Qwen3)
python -m wisent inference-config set --enable-thinking true
Reset to Defaults
python -m wisent inference-config reset

Subcommands

SubcommandDescription
showDisplay current inference configuration
setUpdate inference configuration values
resetReset configuration to default values

Set Arguments

ArgumentTypeDescription
--do-sampleboolEnable sampling (true/false)
--temperaturefloatSampling temperature (e.g., 0.7)
--top-pfloatTop-p (nucleus) sampling (e.g., 0.9)
--top-kintTop-k sampling (e.g., 50)
--max-new-tokensintMax new tokens to generate
--repetition-penaltyfloatRepetition penalty (e.g., 1.0)
--no-repeat-ngram-sizeintNo repeat n-gram size
--enable-thinkingboolEnable thinking mode for Qwen3 models

Related Commands

Stay in the loop. Never miss out.

Subscribe to our newsletter and unlock Wisent insights.