DEDUPED Hot AI repos, turned into use cases Get the Monday 5
197 stars · TypeScript

How to use Osw: what you can do with it

OSW is a local proxy that puts all your LLM channels behind one address and fails over to the next channel when one goes down.

Trended earlier

View repo on GitHub

Third-party repos we don't own. AI-written use cases can be wrong. Disclaimer

What do you want to do?

4 use cases

  • Keep AI coding tools running when one provider fails

    Register several channels in OSW so your client keeps getting answers when one provider breaks.

    View prompt
    Using the OSW repo at yinxulai/osw, configure a local proxy with the channels for <provider-list>, then point <client-name> at http://127.0.0.1:9300/v1. Expected output: a list of model names from `curl http://127.0.0.1:9300/v1/models`. Verify that the models endpoint returns a response before reporting success.

  • Compare provider speed and reliability from request logs

    Use per-request logs and analytics to see which channel answered and how fast it responded.

    View prompt
    Using the OSW repo at yinxulai/osw, review the request logs and analytics for the models in <logical-model-name>. Expected output: a ranked list of models with success rate, TTFT, and TPS. Verify that each figure comes from the Analytics screen before listing it.

  • Send hard requests to a strong model and easy ones to a cheap model

    Use Smart Routing to split traffic into channel groups by model name, client, or request complexity.

    View prompt
    Using the OSW repo at yinxulai/osw, build a Smart Routing policy that sends requests matching <condition> to <channel-group> and everything else to the fallback. Expected output: a saved policy and request log rows showing the chosen channel for each test request. Verify with at least one test request per group before reporting the policy as working.

  • Fix a provider's header or field quirk with rewrite rules

    Add, drop, or pin request fields for one provider using rewrite rules and a built-in test check.

    View prompt
    Using the OSW repo at yinxulai/osw, create a request rewrite rule that <change> for the provider <provider-name>. Expected output: a saved rule and a successful request in Request Logs. Verify that the editor test case passes before saving.

Quick start with Osw

  1. Check you have what it needs: A Mac, Windows PC, or Linux machine that can run the OSW installer; API keys for at least two LLM channels you own; An AI client that lets you set a custom base URL.
  2. Open yinxulai/osw on GitHub and read the README and license.
  3. Download the installer for your platform from the latest release on the OSW GitHub page and install it.
  4. Open Model Management, choose New provider, and enter the name, API key, timeout, and default endpoint for each provider you use.
  5. Add the real model IDs under each provider and tick the protocols each model supports.
  6. Open Logical Models, drag the models into the order you want them tried, and switch off any you want to skip.

These steps come from our top use case, Keep AI coding tools running when one provider fails, which also has a copy-paste agent prompt.

Use this in your AI agent

Ask your agent

Using the Deduped MCP, get the setup steps for yinxulai/osw

Add Deduped to Claude Code once

claude mcp add --transport http deduped https://dedupedai.com/mcp

Cursor, ChatGPT, others

Osw alternatives · Best AI repos for Software and DevOps

About this repo

  • Who it is forDevelopers and power users who run AI clients against several providers, accounts, or models and want one local endpoint that keeps working when a channel fails.
  • Why it is hot197 stars, launched 74 days ago
  • LanguageTypeScript
  • Sourcegithub.com/yinxulai/osw · Back to Hot repos

Requirements

  • A Mac, Windows PC, or Linux machine that can run the OSW installer
  • API keys for at least two LLM channels you own
  • An AI client that lets you set a custom base URL

Add this badge to your README

Maintain Osw? Show readers where to find these use cases. Use cases on Deduped

Markdown 

[![Use cases on Deduped](https://dedupedai.com/badge/yinxulai/osw.svg)](https://dedupedai.com/use-cases/yinxulai-osw/?utm_source=github&utm_medium=badge&utm_campaign=yinxulai-osw)

HTML 

<a href="https://dedupedai.com/use-cases/yinxulai-osw/?utm_source=github&amp;utm_medium=badge&amp;utm_campaign=yinxulai-osw"><img src="https://dedupedai.com/badge/yinxulai/osw.svg" alt="Use cases on Deduped" height="20"></a>

A static image, no tracking script. About the badge

Questions about Osw

What is Osw?

OSW is a local proxy that puts all your LLM channels behind one address and fails over to the next channel when one goes down. It is an open-source project on GitHub: https://github.com/yinxulai/osw.

Who is Osw for?

Developers and power users who run AI clients against several providers, accounts, or models and want one local endpoint that keeps working when a channel fails.

What can I do with Osw?

4 use cases so far, for example: Keep AI coding tools running when one provider fails; Compare provider speed and reliability from request logs; Send hard requests to a strong model and easy ones to a cheap model. Each has steps and a copy-paste agent prompt.

What do I need to run Osw?

A Mac, Windows PC, or Linux machine that can run the OSW installer; API keys for at least two LLM channels you own; An AI client that lets you set a custom base URL.

What are alternatives to Osw?

Other hot AI repos in Software and DevOps: Codex LB Enhanced, Reuse Before Build, Aibuildai LLM Posttrain Agent, FlowX, Codex Guard.

Is Osw safe to run?

Deduped does not own, audit or vouch for it. Read the README and license, run it on your own systems first and review every command before you run it.