Codex
算。子 agent 流程比单 agent 消耗更多 token。
每个子 agent 各自做模型和工具调用。云端任务和本地会话共用一份额度，一条云端对话比一条本地消息用得更多。委派出去的 worker 跟其他 Codex 活动一样计费。并行任务有没有单独的并发上限，官方没写。
官方原文
Because each subagent does its own model and tool work, subagent workflows consume more tokens than comparable single-agent runs.Cloud chats on ChatGPT plans use GPT-5.6 Sol and may use more of your allowance than local messages.These rates apply to supported ChatGPT Work and Codex activity, including local tasks, cloud tasks, automations, code review, auto review, and delegated workers.Claude
算。子 agent、agent 团队、云端并行会话都从同一份额度里扣。
/usage 会把最近的用量归到子 agent、技能、插件和各个 MCP 服务器上。Claude Code 网页版和账号里其他用量共享额度，并行任务按比例消耗更多。agent 团队大约是普通会话 7 倍的 token，因为每个成员都是独立的 Claude 实例。一次大规模扇出可能在会话窗口恢复之前就把周额度耗光。
官方原文
On a Pro, Max, Team, or Enterprise plan, /usage also shows a breakdown of what counts against your plan limits: Attribution: recent usage attributed to skills, subagents, plugins, and individual MCP servers, each shown as a percentage of the total.Rate limits: Claude Code on the web shares rate limits with all other Claude and Claude Code usage within your account. Running multiple tasks in parallel consumes more rate limits proportionately. There is no separate compute charge for the cloud VM.Agent teams use approximately 7x more tokens than standard sessions when teammates run in plan mode, because each teammate maintains its own context window and runs as a separate Claude instance.A single burst of heavy activity, such as a large workflow fanout, can exhaust the weekly allowance before the session window resets.实际影响
- 任务允许的话，把子 agent 路由到更小的模型；主窗口为每一轮子 agent 买单。
- 并行会话在哪都没有单独额度，只是把同一份额度花得更快。
官方原文于 2026-09-15 对照在线文档核对。