
Reasoning Models
The OpenAI reasoning models, the o1, o3
models, DeepSeek R1 or Anthropic Sonet 3.7, are models that are optimized for reasoning tasks.
script({ model: "openai:o1",})
Model Alias
Section titled “Model Alias”The reasoning
and reasoning-small
model aliases are available for reasoning models.
script({ model: "openai:reasoning",})
or
genaiscript run ... -p openai -m reasoning
Reasonong, thinking
Section titled “Reasonong, thinking”GenAIScript automatically extracts the thinking/reasoning content of the LLM responses.
Reasoning effort
Section titled “Reasoning effort”The reasoning effort parameter can be set to low
, medium
, or high
.
- configured with the
reasoningEffort
parameter
script({ model: "openai:o3-mini" reasoningEffort: "high"})
- as a tag to the model name
script({ model: "openai:o3-mini:high",})
For Anthropic Sonnet 3.7, the reasoning efforts are mapped to the following budget_token
values:
- low: 2048
- medium: 4096
- high: 16384
Limitations
Section titled “Limitations”o1-preview
,o1-mini
do not support streamingo1
models do not support tool calling so GenAIScript uses fallback tools.
Advice on prompting
Section titled “Advice on prompting”OpenAI provides an extensive advice on prompting reasoning models.