大模型重试机制隐藏巨额成本，这份研究用“Token膨胀率”重新定义路由策略，为Agent系统省钱指明新方向
arXiv:2608.13571v1 Announce Type: cross Abstract: When a language model fails to answer a query on the first attempt, an agentic system retries, consu…
自托管多代理AI助手，模型无关且能学习偏好，把私有数据变成你的专属智能管家
I didn't want to buy a standalone computer or repurpose a laptop to run constantly so I could maintain a system to sync my LLMs, so I built this. It's…
科大讯飞发布星火Token Factory，用智能路由与全链路治理解决企业大模型规模化落地的运营管理难题
集成多模型并智能路由，自动选模型省钱又高效，登顶DRACO双榜，平均成本仅$0.12
少烧钱、真交付
多模型集成开源AI Agent，智能路由省钱，自组织工作流与可验证编码，登顶榜单对比旗舰Fable5
开源C++代理，智能路由LLM请求，直降70%API成本，自托管更安全。
https://github.com/Cascade-Router/cascade-router Comments URL: https://news.ycombinator.com/item?id=48662106 Points: 2 # Comments: 0
银行每年处理数百万份欺诈报告，这个AI代理通过对话、探询和智能路由，精准分流客户至专家团队，解决人工流程慢和压力大的痛点。
arXiv:2605.16268v1 Announce Type: cross Abstract: Banks receive millions of reports of fraud, scams, and disputed transactions every year, making it c…