An independent project focused on per-request cost visibility across models, gateways, agents, and workflows.
AI costs are typically tracked at the model or monthly level. Real decisions happen one request at a time.
RequestCost is an independent project providing granular visibility into what individual AI requests actually cost — and why.
Request-level cost visibility for teams building and operating AI systems.
Track how routing decisions affect per-request costs across providers and models.
Measure the cost impact of multi-step tasks, tool calls, retries, and fallback chains.
Compare effective cost across models, workloads, and usage patterns in production.
Enable forecasting, budgeting, and chargeback with request-level spend data.
A single AI request can vary significantly in cost depending on multiple operational factors. RequestCost surfaces this layer — giving teams the visibility needed for better decisions.
Core capabilities under development for request-level cost intelligence.
Teams and roles involved in building, operating, and governing AI systems.
RequestCost is currently in early development. Join the waitlist for updates and early access.