What is the Routapse LLM router?

An open-source LLM router. A small router model, Laya or Jev, reads each prompt and decides where it goes: which LLM, which specialist agent, or whether to answer directly without calling a model at all.

How does Routapse use Jev?

Jev is one of the two router models listed in the README, alongside Laya. It classifies each prompt before any big model runs, so you only pay for the power the prompt needs.

Which backends can Routapse route to?

Models can be local through Ollama or hosted from OpenAI, Claude, Gemini or any OpenAI-compatible server, and the README also describes routing to specialist agents, all behind one OpenAI-compatible endpoint.

How do I define routing rules in Routapse?

You describe the routing in plain language in a GUI, and the README does not show the rule format in the digest excerpt. Persian documentation is also available in README.fa.md.

What are the stated benefits of Routapse?

The README's first point is spending less: greetings and lookups go to a small model and only hard prompts reach the expensive one; it also pitches one endpoint in front of many specialist agents.

What do I need to run Routapse?

The badges say Python 3.11+ with a FastAPI backend, a React GUI and an OpenAI-compatible API. Requirements for Jev or Laya access and hosting are not documented in the digest excerpt.

Who built Routapse and under what license?

The GitHub user pooyaphoenix publishes it under the MIT license. The repository was created 2026-10-04 and last pushed 2026-10-10, with 4 stars and 1 fork when the sweep ran.

Jev directory

CommunityJev Integrations & Adapters

Routapse

An open-source LLM router: a small router model, Laya or Jev, reads each prompt and decides which LLM, specialist agent or direct answer it gets, then serves everything through one OpenAI-compatible endpoint. Routing rules are described in plain language in a GUI.

See highlights Open github.com

Stars: 4
Language: Python
License: MIT
Reading time: 2 min
Published by: Community
Author: pooyaphoenix
Last verified: Oct 2026

pooyaphoenix/Routapse

User repository on GitHub

View on GitHub

Cost-efficient LLM routing. Laya and Jev classify each prompt before any big model runs, so you only pay for the power you need. OpenAI-compatible, self-hosted, with a visual router builder.

Forks: 1
Last pushed: Updated Oct 2026

Repo stats from the GitHub API, cached Oct 2026.

Tagscommunity github llm-router gateway

Works forJev for Routing

Used inJev for Software & DevTools

Languages: Python
Platforms: API Browser
Capabilities: Open sourceSelf-hosted

Highlights

Laya or Jev classifies each prompt before any big model runs: which LLM, which specialist agent, or answer directly.
One OpenAI-compatible endpoint fronts local models via Ollama and hosted models from OpenAI, Claude and Gemini.
Routing is described in plain language in a GUI, then saved and called through the endpoint.
Spend less: greetings and lookups go to a small model, and only hard prompts reach the expensive one, per the README.
MIT licensed, Python 3.11+ with a FastAPI backend and React GUI; created 2026-10-04, pushed 2026-10-10.

Watch out

README reviewed 2026-10-17; the digest excerpt ends inside the why-Routapse list, so configuration examples, benchmarks and how Jev or Laya are hosted are not documented here.

Questions about this entry

Short answers drawn from the linked project or page.

Reactions & coverage

Posts, threads, and videos about this entry from around the web.

X: Jev 1.14 is now on OpenRouter

X: Jev lands on the Vercel AI Gateway

X: Jev as an LLM guardrail inside Agentgateway

YouTube: Greg Isenberg and Ryan Vogel on the businesses Jev unlocks, and a path in through the Vercel AI Gateway.

From the Jev guides

Original write-ups on what this entry is for.

Decision gates in agent frameworks: what each integration ships

Oct 16, 2026

How Pydantic AI, Mastra, TanStack AI, DSPy, Spring AI, and LiteLLM expose Jev: where the typed decision sits in the loop, what each integration returns, and which big frameworks are only adjacent.

Jev Integrations & Adapters#agents#frameworks#integrations

Read article

From the Jev AI model community

Posts from builders shipping with Jev right now.

Follow @typesafeai