OpenAI 正在测试一个 Codex 仪表板,可显示哪些聊天正在消耗您的使用配额
Windows build 7868 可以根据五小时、每周和积分使用情况,对最多 100 个 Work 和 Codex 聊天进行排名,并采用了一种新系统,将子代理的活动计入父任务。
By Ryan Merket · Published
RUNTIMEWIRE INVESTIGATION — Scoop
Original reporting by RuntimeWire, based on reverse engineering.
为何重要
Codex 已经显示账户范围的容量和重置时间。新的归因将帮助用户识别导致异常大量消耗的具体 agentic 任务、模型设置和子代理活动。
Reporting record
Finding
OpenAI has built a gated Codex for Windows dashboard that shows users which chats and subagents are consuming their usage limits and credits.
How we verified
Methods: reverse engineering.
RuntimeWire confirmed the page title, intended audience, metrics, rollout controls and chat-ranking behavior in the production client. The accompanying image is a reconstruction using synthetic data.
RuntimeWire identified the feature by comparing consecutive production builds of Codex for Windows and reviewing the newly added interface and supporting application logic.
Tested versions:
- ChatGPT/Codex desktop 26.901.31953 Internal build 7868 Bundled backend 0.153.1 Published September 4
- 2026
Reproduction
RuntimeWire partially reproduced the finding.
File hashes
File: app.asarSize: 299,073,250 bytesSHA-256: 689a59eccd6b4d38f3ddaf202dac05b7cf5ca9cba93b2703f7aa8a44be90b23e
Company response
RuntimeWire requested comment; the company had not responded by publication time.

OpenAI is testing a Codex dashboard that traces five-hour and weekly usage back to individual agentic conversations.
对在 Codex for Windows 构建 7868 中发现的受限 Consumer view 的重建。界面结构和标签基于打包的应用;聊天名称和使用值为合成数据。…
When you purchase through links in our articles, we may earn a small commission. This doesn’t affect our editorial independence.