# Continue + LiteLLM

IDE autocomplete and chat over your models. GitHub Copilot without the seat.

**Replaces:** GitHub Copilot

**Tier:** Easy

## Problem

GitHub Copilot Pro is $10/month per developer, then Pro+ and Max climb the credit ladder. Continue is the OSS IDE extension; LiteLLM already sits in the ChatGPT stack and can serve the same keys.

## Stack

- [Continue](https://continue.dev/) — Open-source IDE Copilot (VS Code, JetBrains)
- [LiteLLM](https://www.litellm.ai/) — Unified proxy over OpenAI, Anthropic, local models

## Estimated SaaS cost saved

| You stop paying | Typical / year | Self-host instead |
|---|---:|---|
| GitHub Copilot Pro (10 seats) | $1,200 | Continue + LiteLLM + your keys |

**About $1,200 / year** on a small team, before you count lock-in.

## Deploy on Railway

Continue runs in the editor. LiteLLM is the Railway piece (already a c17 stack). Point Continue at the LiteLLM OpenAI-compatible URL.

## AGENT_PROMPT

```
Replace GitHub Copilot with Continue pointed at LiteLLM.

- Deploy LiteLLM (see the Open WebUI stack) or reuse that proxy. Continue config uses the OpenAI-compatible endpoint and your model name.
- Document one working VS Code setup. You still pay tokens. You stop paying Copilot seats.
```

## FAQ

### Is this the ChatGPT Team stack?

Same proxy (LiteLLM), different UI. Open WebUI is the chat window. Continue is the editor. One set of keys.

Source: https://c17.io/stacks/continue-copilot