GitHub Repository

Plano is an AI-native proxy and data plane for agentic apps — with built-in orchestration, safety, observability, and smart LLM routing so you stay focused on your agents core logic.

6,585 starsRust

Preference-aware routing for OpenClaw via Plano

Name: Preference-aware routing for OpenClaw via Plano
Availability: InStock
Author: sparacha

by sparacha·Feb 17, 2026·1 point·0 comments

Visit Project View on HN

AI Analysis

●MidNiche GemSolve My Problem

The Take

This stitches Arch-Router into Plano so OpenClaw traffic can be steered to different models by task preference — e.g., cheap k2.5 for calendar/email and Opus 4.6 for heavy app-building — which is a sensible, pragmatic way to shave inference costs without manual swapping. The demo looks usable (config.yaml + README + diagram) but stops at integration; I'd like to see performance/latency comparisons, failure handling and more real-world routing rules before I'd trust it in production.

Post Description

Hey HN!

OpenClaw is notorious about its token usage, and for many the price of Opus 4.6 can be cost prohibitive for personal projects. The usual workaround is “just switch to a cheaper model” (Kimi k2.5, etc.), but then you are accepting a trade off: you either eat a noticeable drop in quality or you end up constantly swapping models back and forth based on usage patterns

I packaged Arch-Router (used by HF: https://x.com/ClementDelangue/status/1979256873669849195) into Plano and now calls from OpenClaw can get automatically routed to the right upstream LLM based on preferences you set. Preference could be anything that you can encapsulate as a task. For e.g. for daily calendar and email work you could redirect calls to k2.5 and for building apps with OpenClaw you could redirect that traffic to Opus 4.6

This hard choice of choosing one model over another goes away with this release.