Parameters Guide

Fine-tune model behavior with parameters.

temperature (0.0-2.0)

Controls randomness:

• 0.0 - Deterministic, focused
• 0.7 - Balanced (default)
• 1.5 - Creative, varied

{"temperature": 0.7}

max_tokens

Maximum tokens to generate:

{"max_tokens": 1000}

top_p (0.0-1.0)

Nucleus sampling:

• 0.1 - Very focused
• 1.0 - Full diversity (default)

{"top_p": 0.9}

Best Practices

Use Case	temperature	top_p	max_tokens
Code	0.2	0.1	2000
Facts	0.3	0.5	500
Creative	1.0	0.9	2000
Chat	0.7	1.0	1000