Running AI workloads without daily cost visibility is risky. This workflow pulls OpenAI usage data and Groq token counts every day — breaking down spend by model, product and date — delivered to email and Telegram.
Breaks down your OpenAI spend by model — GPT-4o, GPT-4o mini, Whisper, DALL·E — so you know exactly what's driving costs.
Pulls Groq token consumption per model — Llama variants — giving you a complete picture of your AI infrastructure spend.
Reports on the last 20 days of usage — long enough to see trends and detect runaway consumption from a single workflow.
Full cost breakdown emailed as a formatted HTML table — by model, by product category and total spend with credit balance.
Concise one-line summary to Telegram: total spend, tokens used, remaining credit balance — at a glance every morning.
Shows your remaining OpenAI credit balance in every report — so you never hit zero mid-workflow unexpectedly.
Daily cron fires. Date range set to last 20 days automatically.
Fetches usage data and credit grants from OpenAI dashboard API — broken down by model and product category.
Groq API queried for token consumption per model for the same period.
JavaScript aggregates costs by model and product. Builds HTML table for email and summary for Telegram.
HTML report emailed. Telegram summary sent. You have full AI cost visibility before your first workflow runs.
If you run AI workflows in production, daily cost visibility is essential — not a nice-to-have. This makes it automatic.
Add LLM cost to your daily infrastructure digest — model-level breakdown lets you optimise routing and reduce costs.
Know your AI bill trajectory every morning. Catch the workflow that's calling GPT-4o when GPT-4o mini would do the job.
We'll connect this to your OpenAI and Groq accounts and configure daily cost reports delivered to your email and Telegram.
Book an Appointment