Chapter
Tokenomics Fundamentals
Connect token consumption to agent behavior, quality, and business outcomes, then find opportunities to improve efficiency with Splunk Agent Observability.
Every prompt, tool call, context, and evaluation adds tokens, latency, and cost. A Gartner study concluded that “Agent models require between 5-30 times more tokens per task than a standard GenAI chatbot[…]”. Token efficiency is becoming the pivotal question of AI adoption.
This workshop explores the Splunk Agent Observability solution, powered by Galileo, to connect token consumption to the traces, spans, quality signals, and business context that produced it.
Workshop Overview #
| Step | Topic | Time |
|---|---|---|
| 0 | How the Lightweight Instrumentation Works | 5 minutes |
| 1 | From Token Counts to Token Value | 2.5 minutes |
| 2 | Three Causes | 2.5 minutes |
| 3 | Gain Visibility | 10 minutes |
| 4 | Choose the Right Model | 10 minutes |
| 5 | Simplify the Workflow | 10 minutes |
| 6 | Connect Spend to Outcomes | 10 minutes |
| 7 | Optimize Evaluation Costs | 5 minutes |
| 8 | Tokenomics - The Big Picture | 5 minutes |
| 9 | Bring It Home | 5 minutes |
A browser is the only requirement for participating in this workshop.
Primary references