# Dari 发布开源路由器模型，通过智能路由和缓存决策在 Terminal-Bench 上以 76 美元成本达到 79.8% 准确率，大幅降低推理费用

- 来源：Chubby
- 发布时间：2026-07-25 02:04
- AIWatch 分数：67
- AIWatch 标记：当日精选
- AIWatch 链接：https://aiwatch.icu/events/evt_01kyaq3ew230v91qa70m08nbxk
- 原文链接：https://x.com/kimmonismus/status/2080715863469224121

## 精选理由

常规快讯，保留列表

## AI 摘要

Dari 发布开源路由器模型，通过智能路由和缓存决策在 Terminal-Bench 上以 76 美元成本达到 79.8% 准确率，大幅降低推理费用。

## 正文

That's the number Dari @daridotdev is launching with today. They open-sourced the router model behind it: a small fine-tuned model that decides which model handles each step, sending most of them to cheaper options and only calling the expensive frontier models when they actually change the result. It factors the cache into that decision too, so it won't switch models when the cached path is the cheaper win. On the same 89-task suite, the frontier setups run into the thousands.

Runs inside Claude Code, Codex, and Pi, you can bring your own Anthropic and OpenAI subscriptions, and the weights are on Hugging Face.