n8n AISIX Model Alias Health Check & Route Probe Report

Know instantly whether your AISIX model aliases are reachable—before users hit errors

This n8n workflow performs an AISIX Model Alias Health Check & Route Probe: it discovers model aliases from your AISIX AI Gateway and sends a small OpenAI-compatible Chat Completions probe to each alias, returning a consolidated report that highlights rate limits, auth failures, policy blocks, and other unhealthy responses.

What this workflow does

• Manual on-demand run via a Manual Trigger.
• Uses a configuration step to set aisix_base_url, a probe_prompt, and a capped max_models to test.
• Calls AISIX /v1/models to fetch the model aliases visible to the configured caller API key.
• If discovery fails or no aliases are available, it returns an attention result to help you detect configuration or connectivity issues quickly.
• Builds a capped queue and sends one non-streaming request per alias to AISIX /v1/chat/completions.
• Classifies each alias result as healthy, rate limited, policy blocked, auth failed, or unhealthy, capturing latency and relevant response headers.
• Aggregates everything into an overall health report and outputs it on either a healthy or attention path.

Use cases

• Pre-flight checks for SaaS operators before deploying routing changes to AISIX.
• Detecting rate limits or policy blocks that can break production chat flows.
• Validating that your caller API key can both list models and call aliases successfully.
• Ongoing monitoring support when troubleshooting “works for some users/models” issues.

Technical details

• n8n nodes: Manual Trigger, Set, Code, Switch, Sticky Note, and HTTP Request.
• Uses an n8n HTTP Header Auth credential to send Authorization: Bearer <your AISIX caller API key>.
• Integrates with AISIX OpenAI-compatible endpoints: /v1/models and /v1/chat/completions (non-streaming).

Best for: n8n users, automation engineers, and teams running AISIX-backed model routing who need fast, reliable alias health visibility.