Claude Code earns its context bill when it operates on a codebase. A reasoning or discussion task — no tool calls, no file edits — runs at a fraction of the cost in a plain chat session.
By Pawan Nayar · from the LLOS AI Learning record
Claude Code earns its context bill when it operates on a codebase.
A reasoning or discussion task — no tool calls, no file edits — runs at a fraction of the cost in a plain chat session.
Claude Code carries a session overhead: system prompt, tool definitions, and accumulated file context all bill as input on every turn. A pure conversation — explaining why output was truncated, reasoning about token accounting — needs none of that. The same exchange in a plain API chat or web session costs orders of magnitude less because there is no agentic scaffolding to haul around and re-bill. If a task touches no files and calls no tools, plain chat is the right instrument.
A developer spends $38 in Claude Code reasoning about token accounting — a conversation that needed zero tool calls. The same exchange in a plain Pro session would cost under $1.
Resolved from this lesson's own record — real definitions, not invented for this page.
More from the same record, pulled by shared topic — not written new for this page.
When Claude Code hit the 200K context limit, it compacted the history — summarising and discarding old turns. This was friction, but it was also a cost governor.
When a background task fails, the AI model must stop. The AI model must explain what broke and why.
An AI model uses various engines to make quiz questions. One engine might use a 'lemma graph' to find related words.
/clear resets the context completely. Use it when a topic changes so the new conversation does not carry the weight of the old one.
Claude Code earns its context bill when it operates on a codebase. A reasoning or discussion task — no tool calls, no file edits — runs at a fraction of the cost in a plain chat session.