LLM nodes across providers, including vision models.
Add these nodes with the pipeline builder: pipeline.add(name="...").<node>(...). Each entry lists the node’s configuration parameters. See the Pipeline reference for add, run, and lifecycle methods.
The data that is sent to the LLM. Add data from other nodes with double curly braces e.g., {{input_0.text}}
str
default:"''"
The system prompt to be used
str
required
Your API key
str
default:"''"
list[str]
default:"[]"
The metadata of the sources used to generate the response
bool
default:"False"
Enable Claude’s built-in web search tool to search the web during response generation
str
default:"''"
str
required
The schema of the JSON response
str
default:"'default'"
Controls the depth of reasoning for GPT-5 models (“none” is only supported on GPT-5.1 variants).
One of: default, high, low, medium, minimal, none
bool
default:"False"
If enabled, the context window will be reduced to fit the model’s maximum context window.
str
default:"'default'"
Controls the verbosity of GPT-5 responses.
One of: default, high, low, medium
SamplingConfig
SafetyConfig
RetryConfig
bool
Whether to enable moderation
bool
Whether to enable PII address
bool
Whether to enable PII cc
bool
Whether to enable PII email
bool
Whether to enable PII name
bool
Whether to enable PII phone
bool
Whether to enable PII ssn
int
The maximum number of retries
int
The maximum amount of input + output tokens the model will take in and generate per run (1 token = 4 characters). Note: different models have different token limits and the workflow will error if the max token is reached.
int
The interval between retries in milliseconds
bool
Enable retrying when the node execution fails
float
The “creativity” of the response - increase the temperature for more creative responses.
float
The “randomness” of the output - higher Top P values increase the randomness
str
default:"''"
The deployment ID for the Azure OpenAI model. This is required when using Azure OpenAI services.