Helicone cost monitoring
CostMon reads cost, tokens, and model straight from Helicone's gateway request log, the layer that knows which user, app, or prompt a call belongs to, and normalizes it into the same daily view as every other provider you connect, so attribution an OpenAI or Anthropic invoice can't give you shows up automatically.
An OpenAI or Anthropic invoice shows one org-level monthly total. It has no idea which of your users, features, or customers drove it. A SaaS company reselling an AI feature at a flat per-seat price can't tell whether a handful of customers are generating most of the spend until someone builds that attribution by hand. Helicone sits in front of every call and can capture that breakdown, but only if requests carry a custom property naming what they belong to, and only if something normalizes that gateway log alongside the rest of the stack instead of leaving it in its own dashboard.
Cost anatomy
Why CostMon
CostMon reads Helicone's gateway request log and turns the custom properties you nominate on the connector into cost dimensions: a workload property by default, plus any extras you configure, such as user or customer. Spend breaks down by the thing you care about instead of one flat provider total.
A normalized per-user view surfaces the accounts driving a disproportionate share of AI spend while paying the same flat price as everyone else. It's visible in days, not after a quarter of margin erosion.
Helicone-attributed spend lands in the same normalized daily table as Anthropic, OpenAI, Bedrock, and Vertex AI, so finance sees one honest AI total instead of a separate gateway dashboard.
CostMon connects with a read-only Helicone API key scoped to request cost and usage data only. It never sees prompts, completions, or account settings.
Getting started
Add a read-only Helicone API key scoped to request cost and usage data. CostMon never sees prompts, completions, or your Helicone account settings.
Per-request rows (cost, tokens, model, and the custom properties you nominated) are pulled and rolled up into a daily view by model, upstream provider, and workload.
Open a daily, per-user or per-workload view of AI spend flowing through Helicone and catch an unprofitable customer or a runaway feature before it erodes margin.
FAQ
Helicone's gateway request log: cost, tokens, and model per request, aggregated into a daily view by model, upstream provider, and workload, plus any additional Helicone custom properties you nominate on the connector as extra cost dimensions.
Two things. Tag requests with Helicone custom properties (the Helicone-Property-* headers) for whatever you want to attribute by, such as a customer id, a feature name, or a prompt version. Then name those properties on the CostMon connector so they become cost dimensions. CostMon carries a workload property by default plus any extras you configure; properties you don't nominate aren't carried through. Untagged or unconfigured traffic still syncs and still counts toward your totals, just without that breakdown.
Helicone's dashboard shows gateway usage on its own. CostMon normalizes that same data alongside AWS, Anthropic, OpenAI, and every other provider you connect, so AI spend routed through Helicone sits in one daily view next to the rest of your stack.
No. The connector reads cost and usage data only, scoped to a read-only Helicone API key. It never requests access to prompt or completion content.
A flat monthly rate per plan, never a cut of your AI spend or the savings CostMon helps you find. Free includes 2 connectors; Professional and Enterprise include unlimited connectors.
Connectors
AWS, GCP, and Azure for cloud; Anthropic, OpenAI, Amazon Bedrock, Vertex AI / Gemini, and Helicone for AI and LLM spend; Snowflake and Databricks for the data cloud; Datadog, GitHub, and Vercel for the rest of the stack. Add CSV import for any tool without a native connector. That's 14 connectors, all normalized into the same unified view.
Connect your first provider and CostMon normalizes it alongside everything else you run. Your team gets one number everyone can check.