Token Spend Tracking & Analytics for GenAI / LLM Costs (Anthropic, OpenAI, etc.)

Customers running GenAI workloads need granular visibility into token consumption and associated costs across models (Anthropic, OpenAI, etc.). Today it is difficult to understand who is consuming tokens, how much, and at what cost — especially when trying to do chargeback, budgeting, or optimization.

We need a dedicated Token Spend capability in Cloudability that allows FinOps and engineering teams to:

• Track token usage (prompt + completion tokens) and cost at a granular level (engineer, team, project, application, model, etc.)
• Create custom views and dashboards focused on token spend (similar to how we have Kubernetes or container views today)
• Drill down into token consumption patterns and trends
• Apply allocation/chargeback rules to token spend
• Set budgets and alerts on token consumption
• Identify waste, anomalies, and optimization opportunities in GenAI spend
  
  This aligns with the emerging Tokenomics / FinOps for AI practice. A good reference for the desired experience is https://www.tokenspend.org/.
  
  Why this is important:
• GenAI spend is growing extremely fast and is often poorly understood - it is the top priority I hear from my customers.
• Engineering teams need visibility to optimize prompts, models, and usage
• FinOps teams need to allocate and charge back these costs accurately
• Without this, Cloudability risks losing relevance for customers with significant AI workloads