Run n8n on Mingles Router
n8n's OpenAI credential has a Base URL field. Point it at Mingles Router once and every AI Agent / OpenAI Chat Model node in every workflow uses it.
Pick your model and paste your key — every command and value on this page updates live.
Endpoint: https://router.mingles.ai/v1 · key stays in your browser ·
no key entered? commands show <your-key>.
Configure once, use everywhere
1. Create the credential
Credentials → Add credential → OpenAI. Paste the API Key and set Base URL to the value below (the field defaults to api.openai.com — change it).
2. Use it in an AI node
In AI Agent or OpenAI Chat Model nodes pick this credential, then enter the model id {{model}} by name (choose “By ID” / expression if the dropdown doesn’t list it).
3. Let it run
A 24/7 inbox-triage or Telegram-post agent is a few hundred thousand tokens a day — this is the workload where the price difference is a bill, not a rounding error.
Exact values
| Credential type | OpenAI |
| Base URL | https://router.mingles.ai/v1 |
| API Key | <your-key> |
| Model (in the node) | deepseek-ai/DeepSeek-V4-Flash-0731 |
Switch model
One credential, all workflows: change the Base URL once to move every automation.
Before you file a bug: read the limits
Reasoning models spend the output budget on internal thinking, output is capped at 8192 tokens, there is no KV cache, and there are no built-in web tools. Most “it broke on Mingles Router” reports are one of these. See Model limits & behavior →
Frequently asked
Do embeddings nodes work? +
No — the endpoint serves chat completions. Keep embeddings on your existing provider and route the chat/agent volume here; the credential split makes that natural.
The model dropdown doesn’t show my model. +
Switch the model field to “By ID” (expression) and type deepseek-ai/DeepSeek-V4-Flash-0731 exactly.