🔄 Авто-синхронизация: из Discussion #1027 каждые 6 часов.

Gonka AI x n8n¶

Автор: @Dankosik · Категория: Show and Tell · Создано: 2026-04-07 12:53 UTC · Обновлено: 2026-04-07 12:53 UTC

📝 Описание¶

GonkaGate x n8n¶

Hi everyone,

We've been working on an n8n package for GonkaGate, and it's finally public.

@gonkagate/n8n-nodes-gonkagate is an open source n8n community node package for people who want to use GonkaGate in self-hosted n8n without piecing together generic OpenAI-compatible nodes by hand.

At a glance¶

Item	Value
Package	`@gonkagate/n8n-nodes-gonkagate`
Root node	`GonkaGate`
AI model node	`GonkaGate Chat Model`
Credential	`GonkaGate API`
Best fit today	Self-hosted `n8n`
Repo	github.com/GonkaGate/n8n-nodes-gonkagate
npm	npmjs.com/package/@gonkagate/n8n-nodes-gonkagate

Why this exists¶

The fallback OpenAI-compatible route works. We've used it too. It just makes people do more translation work than they should before the first request.

If you want to use GonkaGate in n8n, you probably should not have to start with base URLs, custom credential wiring, and a guess about which stock node is the least painful in your current n8n version. We wanted a GonkaGate-first path that feels shorter and clearer.

We also did not want to pretend the package does more than it does. It should be clear about what GonkaGate supports today, while still leaving room to grow later without renaming everything around one endpoint.

What you get¶

the root node GonkaGate
the additive AI model node GonkaGate Chat Model
the shared credential GonkaGate API

Current API scope¶

[!NOTE] The package currently targets the public GonkaGate base URL https://api.gonkagate.com/v1.

Right now that means these endpoints:

GET /v1/models

POST /v1/chat/completions

In practice, the package can do the following today:

install both GonkaGate and GonkaGate Chat Model from one package
reuse one GonkaGate API credential across both
run List Models
run non-streaming root-node Chat Completion
use GonkaGate Chat Model in n8n AI workflows
load models from GET /v1/models
fall back to manual Model ID entry if the live list is empty or unavailable

That last part was a deliberate choice. Live model discovery is nice when it works, but manual Model ID still matters.

Quick start¶

If you already run self-hosted n8n, the shortest install path is still the Community Nodes UI:

@gonkagate/n8n-nodes-gonkagate

Then:

Create GonkaGate API.
Add the root GonkaGate node.
Run List Models.
Switch to Chat Completion for the first real request.

What we're not claiming¶

[!IMPORTANT] A few things we are deliberately not claiming yet:

blanket support across all n8n versions
n8n Cloud availability
verified-node eligibility
/v1/responses support
broad live validation for every AI-agent or tool-calling-heavy workflow shape

For the exact support posture, see the compatibility matrix and known limitations.