AI article
I capped my coding agent's tool output to save tokens. It used 2.65 times as many.
Community description: Claude Code token usage, measured: a 12,000-character cap on tool output took one answer from 267k to 708k billed input tokens. Why, and what 252 runs showed.
Dev.to | Nov 2, 2026 | Rodion Kazennov
Automated excerpt
I wanted to know whether a code graph makes Claude Code, Anthropic's coding agent, cheaper than plain search. My very first graph run looked six times cheaper than grep. The graph never beat Read and Grep on tokens.
Selected automatically from source text; not independently written or fact-checked. Read the original for full context.