<!-- GENERATED BY scripts/gen_markdown_twins.py; source: guides/n8n/index.html -->

# Use n8n with Chinese AI models through ChinaAPI.

> Configure n8n OpenAI credentials with ChinaAPI's custom base URL for DeepSeek, Kimi, Qwen, GLM, and MiniMax in AI Agent and LLM nodes.

- Canonical page: https://chinaapi-ru.com/guides/n8n/
- Human-readable page: https://chinaapi-ru.com/guides/n8n/
- Live pricing: https://dash.chinaapi-ru.com/pricing?utm_source=chinaapi&utm_medium=markdown&utm_campaign=n8n

Create OpenAI-compatible credentials in n8n, set ChinaAPI as the custom base URL, and use official-price Chinese models inside AI Agent and LLM nodes with one key.

[Start free — $2 credit](https://dash.chinaapi-ru.com/register?utm_source=chinaapi&utm_medium=web&utm_campaign=guide-n8n)

[Live model pricing](https://dash.chinaapi-ru.com/pricing)

## What to prepare.

### n8n instance

Use n8n Cloud or a self-hosted n8n instance with AI Agent, OpenAI, Chat OpenAI, or LM OpenAI nodes available.

### ChinaAPI key

Register, open Dashboard -> Console -> Tokens, and copy an `sk-...` key.

## Create OpenAI credentials with a custom base URL.

- In n8n, open Credentials and create new OpenAI credentials, or create them from an OpenAI/Chat OpenAI/LM OpenAI node.

- Choose API key authentication.

- Paste your ChinaAPI token into **API Key**. Leave **Organization ID** blank unless your n8n build requires a value.

- Set the custom base URL, base URL, or API endpoint field to `https://api.chinaapi-ru.com/v1`. Some n8n versions show this under additional options.

- Save the credential.

- In an AI Agent, Chat Model, LM OpenAI, or OpenAI node, select the credential and enter an exact model ID such as `deepseek-v4-flash`.

## Use a simple AI node prompt.

```
Return JSON only:
{"status":"connected","provider":"ChinaAPI","tool":"n8n"}
```

If the node supports manual model entry, use `deepseek-v4-flash` for the first run and keep temperature low.

## Good n8n defaults.

For current model details and rate evidence, review [DeepSeek-V4 Flash](https://chinaapi-ru.com/models/deepseek-v4-flash/) and [Qwen3.6 Flash](https://chinaapi-ru.com/models/qwen3.6-flash/) before setting a production workflow default.

### `deepseek-v4-flash`

Fast workflow automation, extraction, and routing logic.

### `qwen3.6-flash`

Low-latency model for high-volume agent steps and summaries.

### `MiniMax-M2.7`

Efficient chat model for routine workflow nodes.

### `glm-5.1`

Reasoning and tool-friendly option for more complex automations.

### `kimi-k2.6`

Useful for long-context workflow prompts and document-heavy tasks.

## Common n8n issues.

Thinking models such as `qwen3.7-max` can generate reasoning tokens before returning final output. Enable streaming if the node supports it or raise the n8n node/execution timeout for long tasks.

**Model not found:** check the exact model ID on [live pricing](https://dash.chinaapi-ru.com/pricing). If the n8n node has a model dropdown only, look for a manual/custom model option.

**401:** confirm your OpenAI credential uses the `sk-...` ChinaAPI key and the request is sent with a bearer authorization header. Re-save the credential after changing the base URL.

## Run n8n AI workflows with gateway-synchronized token pricing.

Rates may follow providers' official China list prices where configured. Signup includes $2 free credit. Use live pricing to confirm model IDs before production workflows.

[Register — $2 credit](https://dash.chinaapi-ru.com/register?utm_source=chinaapi&utm_medium=web&utm_campaign=guide-n8n)

[Live pricing](https://dash.chinaapi-ru.com/pricing)

---

Markdown twin of https://chinaapi-ru.com/guides/n8n/ — generated from its canonical HTML by scripts/gen_markdown_twins.py. Full site reference: https://chinaapi-ru.com/llms-full.txt