thinking

AI Agent Extended Thinking Budgets Module for PraisonAI Agents. Provides configurable thinking budgets for LLM reasoning:

Token budgets for extended thinking
Time budgets for reasoning
Adaptive budget allocation
Budget tracking and reporting

Zero Performance Impact:

All imports are lazy loaded via getattr
Budgets only applied when configured
No overhead when not in use

Usage:

from praisonaiagents import Agent
from praisonaiagents.thinking import ThinkingBudget

# Create a thinking budget (helper for computing per-task token limits)
budget = ThinkingBudget(
    max_tokens=16000,
    max_time_seconds=60,
    adaptive=True
)

# Compute an adaptive token budget for a given task complexity (0.0-1.0)
tokens = budget.get_tokens_for_complexity(0.8)

# `agent.thinking_budget` is a backward-compatible alias for the unified
# `reasoning_effort` control. Prefer the graded level directly, which the
# core request pipeline translates to each provider's native parameter
# (OpenAI/xAI `reasoning_effort`, Anthropic/Gemini extended-thinking budget):
agent = Agent(instructions="...", reasoning_effort="high")

Import

from praisonaiagents import thinking