Apps · Productivity
n8n
Workflow automation with LLM nodes.
Tokens / 7d
47B
Routing directive
auto:cheap
On our metal
33%
Licence
Open source
n8n workflows call models on a schedule or a webhook — classification, extraction, summarization. Volume is steady and predictable, and almost none of it needs a frontier model.
How it connects
- OpenAI credential pointed at the gateway base URL.
- Batch-friendly routing with generous retry budgets.
- Structured output mode for extraction nodes.
Traffic pattern
Flat, scheduled load. Ideal candidate for batch pricing and the cheapest capable model.
gemini-3.6-flashgpt-4.1-miniclaude-haiku-4.5
Point it at the gateway
// n8n → Credentials → OpenAI Base URL: https://api.llmcloud.ai/v1 Model: auto:cheap