--max-tokens caps the model’s output length. When you pass it explicitly, PraisonAI honours it verbatim — even when the value equals the built-in default.
Quick Start
1
Set a budget from your agent
2
Cap output from the CLI
Behaviour change (PR #5360 / issue #5358). Before this fix,
--max-tokens 16000 was silently dropped because the resolver could not distinguish “user typed 16000” from “argparse filled in the default”. The parser now marks explicitness with a sentinel at parse time, so --max-tokens 16000 behaves identically to --max-tokens 15999. Any prior workaround like --max-tokens 16001 is no longer needed.How It Works
The parser tracks whether you actually typed the flag, so an explicit value always wins — regardless of what it equals.Precedence
The most specific source that you actually set wins.Usage
Common Patterns
Short Response
Long-form Content
With Research
Token Limits by Model
Best Practices
Type the flag when you mean it
Type the flag when you mean it
An explicit
--max-tokens N always wins — including --max-tokens 16000. You no longer need workarounds like --max-tokens 16001.Let YAML own per-agent budgets
Let YAML own per-agent budgets
Omit
--max-tokens to let each agent’s max_tokens: in YAML apply. The CLI default only fills in when nothing else set a budget.Stay under the model cap
Stay under the model cap
Values above the model’s output limit are capped automatically, so match your budget to the model you run.
Related
Agent Max Budget
Set per-agent token and cost budgets in code.
Model-aware clamp (#4124)
Tracking issue for clamping budgets to each model’s cap.