\n
\n Coding assistants like Aider, Cline, and Continue all speak the OpenAI wire protocol \u2014 point them at a base_url, give them an API key, done. That makes swapping in a different LLM backend trivial… if that backend uses Authorization: Bearer.<\/p>\n
The flat-priced, auto-routing API I’d been using doesn’t. It’s distributed through RapidAPI, which authenticates with an X-RapidAPI-Key header instead of Bearer. So I couldn’t just drop it into Aider. The fix turned out to be ~120 lines, so I open-sourced it.<\/p>\n
modelis-openai<\/p>\n
A zero-dependency local proxy (MIT, Node 18+). It listens on 127.0.0.1, speaks plain OpenAI, rewrites the auth header, and forwards to the upstream gateway. Streaming (stream: true) is piped straight through, so token-by-token output works exactly as with the OpenAI API.<\/p>\n
your tool \u2500\u2500OpenAI(Bearer)\u2500\u2500\u25b6 modelis-openai (localhost) \u2500\u2500X-RapidAPI-Key\u2500\u2500\u25b6 upstream \u2500\u2500\u25b6 best model<\/p>\n
Enter fullscreen mode<\/p>\n
Exit fullscreen mode<\/p>\n
Quickstart<\/p>\n
npx modelis-openai<\/p>\n
Enter fullscreen mode<\/p>\n
Exit fullscreen mode<\/p>\n
Then point any OpenAI-compatible tool at it:<\/p>\n
Setting
\nValue<\/p>\n
Base URL
\nhttp:\/\/127.0.0.1:8787\/v1<\/p>\n
API key
\nyour RapidAPI key<\/p>\n
Model
\nmodelis-auto<\/p>\n
Drop it into your tool<\/p>\n
Aider<\/p>\n
export OPENAI_API_BASE=http:\/\/127.0.0.1:8787\/v1
\nexport OPENAI_API_KEY=
\naider –model openai\/modelis-auto<\/p>\n
Enter fullscreen mode<\/p>\n
Exit fullscreen mode<\/p>\n
Cline \/ Roo Code \u2014 API Provider OpenAI Compatible, Base URL http:\/\/127.0.0.1:8787\/v1, Model ID modelis-auto.<\/p>\n
Continue (~\/.continue\/config.yaml)<\/p>\n
models:
\n – name: Modelis
\n provider: openai
\n model: modelis-auto
\n apiBase: http:\/\/127.0.0.1:8787\/v1
\n apiKey: <\/p>\n
Enter fullscreen mode<\/p>\n
Exit fullscreen mode<\/p>\n
Any OpenAI SDK<\/p>\n
from openai import OpenAI
\nclient = OpenAI(base_url=”http:\/\/127.0.0.1:8787\/v1″, api_key=””)
\nprint(client.chat.completions.create(
\n model=”modelis-auto”,
\n messages=({“role”: “user”, “content”: “Hello”}),
\n).choices(0).message.content)<\/p>\n
Enter fullscreen mode<\/p>\n
Exit fullscreen mode<\/p>\n
How it works<\/p>\n
Reads the key from Authorization: Bearer (or MODELIS_RAPIDAPI_KEY).
\nRewrites the request model to modelis-auto (configurable).
\nForwards to the RapidAPI gateway with X-RapidAPI-Key \/ X-RapidAPI-Host.
\nRelays the response \u2014 including SSE streams and rate-limit headers \u2014 unchanged.<\/p>\n
It also answers GET \/v1\/models and GET \/health so tools that probe on startup don’t choke.<\/p>\n
Honest notes<\/p>\n
It routes to a paid API (there’s a free tier to start). The point of the proxy is to remove the integration friction, not to give anything away.<\/p>\n
Cursor isn’t supported \u2014 it sends requests from its own servers, so a localhost endpoint can’t be reached. This is for tools that call the API from your machine.<\/p>\n
Links<\/p>\n
If you try it in a tool I didn’t list, I’d love to hear how it goes.<\/p>\n