ai-cost-reduction 开源项目推荐（11 个）

NadirRouter /NadirClaw

开源的大型语言模型路由器和人工智能成本优化器。简单提示自动路由到廉价/本地模型，复杂提示自动路由到高级模型。可直接加入的 OpenAI 兼容 Claude Code、Codex、Cursor、OpenClaw 代理。节省40%-70%的AI API成本。自架，没有中间人。

ColorsCodeView78 /AI-Token-Saver-2026---Cut-Token-Costs-by-75-Cloud-Code-Optimization-Tool

Professional AI token optimization software that reduces token usage by up to 75% for developers and teams. Boost efficiency and lower AI expenses in 2026.

anshmajumdar121 /context-optimizer

Reduce Claude AI token consumption. Zero install to start. Python 3.7+ for auto-manifest generation.

budgetguard-ai /budgetguard-core

A FinOps control plane for AI APIs - Drop-in API gateway that enforces hard budgets, rate limits, and custom policy checks before any request reaches AI providers. Stop surprise AI bills with multi-tenant budget controls, complete audit trails, and sub-20ms latency overhead

blackwell-systems /gcf-codex-plugin

Codex plugin for GCF Proxy. Save 71% on MCP tool call tokens.

blackwell-systems /gcf-claude-plugin

Claude Code plugin for GCF Proxy. Save 71% on MCP tool call tokens.

adaptiq-ai /adaptiq

Adaptive Optimization Framework for AI Agents using Reinforcement Learning. Automatically optimize agent behaviors, reduce costs by 30%, and improve performance through Q-learning and data-driven prompt engineering.

j1s4nn /jevXagent

When Jev meets LLM — Transparent proxy that makes AI coding agents 3× faster and 80% cheaper. Cut response times from 2.4s to 890ms with Line J architecture.

GlazePrivatePray46 /AI-Efficiency-Boost-2026---Cut-Token-Costs-by-75-Smart-Optimization-Tool

Powerful AI efficiency tool that reduces token usage by up to 75% for cloud code and LLM applications. Ideal for developers looking to maximize performance while minimizing costs in 2026.

FutureTaleHinge /Cloud-Code-Optimizer-2026---Reduce-Token-Usage-by-75-AI-Efficiency-Tool

Advanced cloud code optimization tool that dramatically reduces token consumption by up to 75% for AI and LLM applications. Perfect for developers and teams seeking maximum efficiency in 2026.

Alpha-Park /genpark-context-sliding-window-entropy-compressor-skill

Computes per-turn Shannon information entropy and lexical density across dialogue trajectories to prune low-information filler and compress context within strict token budgets | GenPark: https://www.genpark.ai | LinkedIn: https://www.llinkedin.com/company/92955550