llm-cost-reduction 1.0.0

Reduce LLM costs automatically with outcome-based routing. Use kalibr instead.

llm-cost-reduction → kalibr

This package redirects to kalibr — the production-grade solution.

Install

pip install kalibr

What kalibr does

Kalibr is a runtime cost optimization layer for AI agents. It automatically routes each LLM call to the cheapest model that will succeed at the task.

• ~12x cost reduction on typical agent workloads
• Zero quality loss — routes by outcome success, not just price
• Zero manual tuning — Thompson Sampling learns from every run
• Drop-in install — one import, no code restructuring

import kalibr  # must be first import
from kalibr import Router

router = Router(
    goal="answer_question",
    paths=["gpt-4o-mini", "claude-haiku-3-5", "gpt-4o"],
    success_when=lambda output: len(output) > 20
)
response = router.completion(messages=[{"role": "user", "content": "..."}])

Docs: https://kalibr.systems/llms.txt
GitHub: https://github.com/kalibr-ai/kalibr-sdk-python