# Not Diamond Code 发布，推理路由层降本 20%+

- 来源：Rohan Paul (@rohanpaul_ai)
- 发布时间：2026-08-05 01:18
- AIHOT 分数：44
- AIHOT 链接：https://aihot.virxact.com/items/cmsexviqv1akhro2ecov2yjjq
- 原文链接：https://x.com/rohanpaul_ai/status/2084690482312155509

## AI 摘要

Not Diamond 发布 Not Diamond Code，一个为长时程编码智能体设计的模型路由层，可在每一步切换模型。官方称其可将成本降低 20% 以上且不损失整体任务质量，并兼容 Claude Code 等网关。在基准测试中，混合路由策略相比 Anthropic 单一策略性能提升 3.6%，节省成本从 39% 增至 66%。

## 正文

So much brilliant work is coming on the routing layer of inference.

Not Diamond just announced Not Diamond Code， a routing layer that lets a coding agent switch between models at every step.

claims this cuts costs by 20% or more without reducing overall task quality.

The point is long-horizon coding makes model routing much harder than picking the cheapest model for each prompt.

A cheap model may handle the current turn， but switching to it can discard a warm KV cache or leave later steps with weaker context. Not Diamond Code instead predicts the future cost and reward of each model and reasoning-effort choice across the remaining agent trajectory.

e.g. the below graph is taken from their official blog， show that how different models are better at different parts of a coding task， and a router that can choose among them can outperform an Anthropic-only strategy while spending less.
In this benchmark， the mixed router improved performance by 3.6% and increased savings from 39% to 66%.

### 引用推文

> Tomas Hernando Kofman：Today we're announcing Not Diamond Code, the world's most powerful intelligent model router for long-horizon coding agents. Not Diamond works with any gateway o...