Library · evaluation

Token budgets change agent capability

Added 2026-07-14 · Last verified 2026-07-14

UK AI Security Institute evaluations found that increasing an agent's token budget from 1 million to 10 million tokens improved success on software-engineering tasks by roughly 25 percent. Some long-horizon cybersecurity tasks can require budgets near 50 million tokens. A weak result may therefore reflect an exhausted reasoning budget rather than a hard capability limit.

Useful?

Read the source ↗


Source: The Agent Loop — agent-loop.xyz. Curated in the open knowledge base (CC BY 4.0).

Ask a question about this ↗