Keep Flowise visual, make model spend visible.
Flowise chains can call an LLM many times per user request. ProxyLLM sits behind the OpenAI node, serves those calls through Codex Hosted on your ChatGPT subscription, and holds every flow to a hard budget.
$129/month SaaS. Bring your own model keys. No inference markup.
Three steps to connect.
Create a Flowise key
Generate a scoped ProxyLLM key for each Flowise deployment, bot, or team.
Use ChatOpenAI custom base path
In the OpenAI or ChatOpenAI credential, set the API key to your ProxyLLM key and the base path to https://api.proxyllm.ai/v1.
Run flows on your subscription
Keep the canvas intact. OpenAI-model steps bill to your ChatGPT plan through Codex Hosted and fall back to your API key at a plan limit. Every call lands in the request log.
Use the same ChatOpenAI node.
Change the credential, not the graph.
OpenAI API Key: pk_live_...
Base Path: https://api.proxyllm.ai/v1
Model Name: gpt-4o-miniRun your AI workloads on your ChatGPT subscription.
ProxyLLM runs OpenAI's Codex for you, signed in with your own ChatGPT account. Your apps call one OpenAI-compatible endpoint and the work bills to your flat plan instead of per-token API pricing.
Cap the flows that call themselves.
Agentic Flowise graphs retry and loop. ProxyLLM gives each one a capped key, and the loops bill to a flat subscription instead of a metered API.