OpenAILLM
OpenAI large language model (LLM).Constructor
OpenAI API key. Defaults to the
OPENAI_API_KEY environment variable.Model id. Default
'gpt-5.4-nano'.Sampling temperature; higher is more random. Default
0.7.Maximum number of tokens to generate. Default
1024.Nucleus sampling probability mass (0-1). Default
null.Penalizes new tokens by their frequency so far (-2.0 to 2.0). Default
null.Penalizes new tokens by whether they appear so far (-2.0 to 2.0). Default
null.Seed for deterministic sampling. Default
null.Output format specification, e.g.
{ type: 'json_object' }. Default null.Controls tool selection, e.g.
'none', 'auto', 'required', or a tool name. Default null.Whether to allow multiple tool calls to run in parallel. Default
null.Stop sequence(s) at which generation halts. Default
null.Stable identifier for the end user, used for abuse monitoring. Default
null.Reasoning effort for reasoning-capable models, e.g.
'none', 'low', 'medium', 'high'. Default null.Response verbosity for capable models, e.g.
'low', 'medium', 'high'. Default null.Stream tokens from the model. Default
false.Override the realtime websocket URL. Default
null.Whether OpenAI may store the request/response for later retrieval. Default
true.