# Baseten — Introducing Baseten Router: a Pareto-optimal model router for coding tasks

- Company: Baseten (baseten.co)
- Announced: 2026-11-04
- Category: developer-tool-release
- Coverage: not counted
- Announcement: yes
- Group: announcements
- Source: https://www.baseten.co/blog/introducing-baseten-router
- Record: https://forck.live/items/32761-introducing-baseten-router-a-pareto-optimal-model-router-for-coding-tasks
- Subject: Baseten
- Models affected: Opus 5.5, GLM 5.3 Flash

Baseten has built a router for Baseten Grid that sends each coding task to either a frontier closed-weight model or a cheaper open-weight one, so customers pay frontier prices only when a task needs them. In its tests, Baseten Router came within 0.8 percentage points of Opus 5.5's pass rate at 72% lower cost, measured on a test set spanning DeepSWE, SWE Atlas QnA and Terminal-Bench. The router decides between Opus 5.5 and GLM 5.3 Flash, which Baseten says is a strong coding model roughly 80% cheaper than Opus 5.5 across relevant workloads. It scores each model by predicted success rate minus a penalty proportional to predicted cost, and a configurable weight sets the trade-off along the cost-performance Pareto frontier. Baseten says it is the only router it evaluated, among commercial routers including those from Fireworks, Together AI and Applied Compute, that beats random routing at the same cost by a statistically significant margin. Baseten Router is part of Baseten Grid, now in public preview, and the company says it is working with design partners to specialize it for production coding workloads.

## Evidence

Verbatim from https://www.baseten.co/blog/introducing-baseten-router:

> Baseten Router sends each coding task to the right model, open or closed, so you pay frontier prices only when the task needs it. In benchmarks, it came within 0.8 percentage points of Opus 5.5 at 72% lower cost. It’s the only router we tested that beat random routing by a statistically significant margin. Baseten Router is part of Baseten Grid, now in public preview.
> 
> Today, we’re sharing our initial results from training an intelligent model router for Baseten Grid.
> 
> Not every coding task needs the most capable model. As AI coding spend continues to grow rapidly [1] and cheaper open-weight models close the capability gap [2] , choosing the right model for each task becomes increasingly valuable. Baseten Router chooses between models based on their expected performance and cost for each coding task. The goal is frontier performance while paying frontier prices only when the task calls for it.
> 
> Baseten Router cuts costs by 72% compared to using Opus 5.5 for every task while remaining within 0.8 percentage points of its pass rate, measured on a test set spanning DeepSWE, SWE Atlas QnA, and Terminal-Bench.

---

Record: https://forck.live/items/32761-introducing-baseten-router-a-pareto-optimal-model-router-for-coding-tasks
Catalogue: https://forck.live/llms.txt
Current issue: https://forck.live/feed.md
