Line of inquiry
Inquiring lines›How does AI reshape human reasonin…›How do training data and procedure…›this line of inquiry
What drives capability and cost efficiency in agent systems?
A broader line of inquiry — a family of 37 specific questions the research asks around this. Follow one into its inquiring-line page, or move sideways to a related line below.
Questions in this line of inquiry 37
Specific inquiring lines the field asks around this — ordered from the most general framing down to the most specific angle.
- Does upgrading model capability improve token efficiency in agentic systems?
- Should production agents execute one tool or multiple tools per invocation?
- Why do production agents depend more on their surrounding pipeline than the model?
- How do tool invocations drive agentic cost beyond token consumption?
- Which layer of agent systems creates the largest capability gains in practice?
- Can we design efficient agents by targeting constraints directly?
- Does effective feedback compute matter more than raw token expenditure for agent scaling?
- How do agents discover and select which tools to invoke?
- What separates good workflow design from poor workflow design?
- How does the execution layer constrain agent performance in tool use?
- What metrics replace throughput per token for agent deployment?
- Do multi-agent systems justify their token costs with genuine quality gains?
- When should you optimize agent behavior versus tool performance separately?
- Why does capability discovery become the bottleneck in large agent systems?
- How will the agent economy reshape compute infrastructure design?
- Why do 85 percent of production agents avoid third-party frameworks?
- Which ecosystem conditions matter most for agent deployment success?
- How do cache-dominant workflows change the marginal cost of agent tasks?
- Should agent capability be optimized separately from general capability?
- Why do multi-agent systems use 15 times more tokens than chat interactions?
- Why do production AI agents deliberately stay simple and avoid frameworks?
- How do planning and memory compress agentic system costs?
- How much does external API latency dominate total agent execution cost?
- Why has agent research prioritized policy over world model development?
- How much does agent performance depend on demonstration quantity versus curation quality?
- Why do rigid orchestration frameworks fail where generative environment specifications succeed?
- When is 15x token overhead actually worth the compute cost?
- What ecosystem conditions make agent attention markets viable?
- Which agent architectures consistently outperform base models on hard prediction questions?
- Can two agents with identical token counts produce vastly different outputs?
- What happens when tools compete for agent invocation rather than human clicks?
- What five ecosystem conditions must coordination governance and evidence actually satisfy?
- Why do APIs outperform UIs for agent task completion?
- What capability threshold do agents need to self-organize effectively?
- What structural constraints produce recursion costs in agentic systems?
- How do controllable simulators compare to population-level agent simulation approaches?
- Why does partial observability require interaction instead of better reasoning?