# Continue

Local VS Code/JetBrains extension or cn CLI; separate from GitHub Copilot.

Source: https://zrelay.net/docs/agents/coding-agents/continue/

## Manual setup

1. Open Local Config or ~/.continue/config.yaml (Windows: %USERPROFILE%\.continue\config.yaml). Append the shown model to the existing models list. Include the name/version/schema/models header only when creating a new file.

2. Enter ZRELAY_API_KEY locally in ~/.continue/.env, preserving other entries and restricting access to your account. Never put its value in YAML or Git. Workspace .env entries take precedence. IDE extensions need this file; a shell export alone is insufficient. The CLI can also inherit process environment.

3. Restart the IDE, choose Local Config > Zrelay GPT and Chat mode. CLI: check cn --version and cn --help; launch cn --config ~/.continue/config.yaml, or cn --config "$env:USERPROFILE\.continue\config.yaml" in PowerShell. Inspect errors without a prompt.

4. Use a GPT key for the enabled gpt-6.1-sol Chat Completions route in Flex/Prime. For a separate Claude entry use provider: anthropic, its exact enabled model/key, apiBase: https://api.zrelay.net/v1/ and maxTokens: 4096. Continue adds messages; the generic Anthropic SDK root base is wrong here.

## Configuration and commands

```
name: Local Config
version: 1.0.0
schema: v1
models:
  - name: Zrelay GPT
    provider: openai
    model: gpt-6.1-sol
    apiBase: https://api.zrelay.net/v1/
    apiKey: ${{ secrets.ZRELAY_API_KEY }}
    roles: [chat]
    defaultCompletionOptions:
      maxTokens: 4096
```

## Alternative: ask an AI assistant

Copy this complete prompt into your coding assistant. It describes configuration work; installation and paid tests need your authorization.

```
Configure Continue locally for Zrelay. Detect my OS, installed IDE/CLI variant and version. Consult https://docs.continue.dev/reference, https://docs.continue.dev/faqs, https://docs.continue.dev/cli/configuration and https://keys.zrelay.net/api/catalog. Preserve all existing settings, models and credentials; inspect configurations with secrets redacted. Ask before installing software or making paid inference requests.
Add one uniquely named model to the active local config.yaml: provider=openai, model=gpt-6.1-sol, apiBase=https://api.zrelay.net/v1/, apiKey=${{ secrets.ZRELAY_API_KEY }}, roles=[chat], defaultCompletionOptions.maxTokens=4096. Confirm that my selected GPT pool enables that model's Chat Completions route and current output limit. For a new file only, add name, version=1.0.0, schema=v1 and models. Guide me to enter the key locally in ~/.continue/.env; CLI can alternatively inherit ZRELAY_API_KEY. Never request, print, copy, log or commit the key. Check workspace secret precedence without exposing values. Explain IDE restart/model selection or the CLI --config path. Validate schema/configuration without inference; report actual checks and limitations. Do not claim autocomplete, editing, tools or streaming have been verified.
```

## Limits and verification

The initial roles list is chat only. edit/apply, autocomplete, embeddings and reranking are separate features, not implied by provider support. Streaming and tools remain untested.

These recipes were reviewed against documentation and public catalog data. No client installation, authenticated request, paid inference, streaming or tool loop was executed. Validate configuration first; a later short text test does not establish complete agent compatibility.

Recheck the [live catalog](https://keys.zrelay.net/api/catalog) and your key’s provider/pool. The reviewed examples have 128000 input and 4096 output maxima unless a client limitation above prevents controlling the request. Never print, log or commit credentials.

## Official sources

Documentation checked 2026-11-09; mutable sources can differ from your installed version.

[https://docs.continue.dev/reference](https://docs.continue.dev/reference)

[https://docs.continue.dev/faqs](https://docs.continue.dev/faqs)

[https://docs.continue.dev/cli/configuration](https://docs.continue.dev/cli/configuration)

[https://keys.zrelay.net/api/catalog](https://keys.zrelay.net/api/catalog)

[https://docs.continue.dev/customize/model-providers/top-level/openai](https://docs.continue.dev/customize/model-providers/top-level/openai)

[https://docs.continue.dev/customize/model-providers/top-level/anthropic](https://docs.continue.dev/customize/model-providers/top-level/anthropic)

[https://docs.continue.dev/customize/deep-dives/configuration](https://docs.continue.dev/customize/deep-dives/configuration)

[https://raw.githubusercontent.com/continuedev/continue/main/packages/openai-adapters/src/apis/OpenAI.ts](https://raw.githubusercontent.com/continuedev/continue/main/packages/openai-adapters/src/apis/OpenAI.ts)

[https://raw.githubusercontent.com/continuedev/continue/main/core/llm/llms/Anthropic.ts](https://raw.githubusercontent.com/continuedev/continue/main/core/llm/llms/Anthropic.ts)

[https://raw.githubusercontent.com/continuedev/continue/main/packages/openai-adapters/src/apis/Anthropic.ts](https://raw.githubusercontent.com/continuedev/continue/main/packages/openai-adapters/src/apis/Anthropic.ts)

[https://keys.zrelay.net/api/catalog](https://keys.zrelay.net/api/catalog)