-
-
Notifications
You must be signed in to change notification settings - Fork 122
Open
Description
Is there a way to specify the parameters verbosity and effort for the model gpt-5-nano-2025-08-07?
I tried to add them like this in my config but i'm not sure they are taken into account:
{
name = "gpt5o",
chat = true,
command = true,
model = { model = "gpt-5-nano-2025-08-07", verbosity = "low", effort = "low" },
system_prompt = Personas.developer
},
The generated query looks like this:
{
"stream": true,
"top_p": 1,
"model": "gpt-5-nano-2025-08-07",
"temperature": 1,
"max_completion_tokens": 4096,
"messages": [
{
"role": "system",
"content": "You are a developer...."
},
{
"role": "user",
"content": "write a hello world function in python"
},
{
"role": "assistant",
"content": "```python\ndef hello_world():\n return \"Hello, World!\"\n\nif __name__ == \"__main__\":\n print(hello_world())\n```"
},
{
"role": "user",
"content": "Summarize the topic of our conversation above in two or three words. Respond only with those words."
}
]
}
Metadata
Metadata
Assignees
Labels
No labels