ai-zero-token 1.0.8 → 1.0.10

package/README.md

@@ -21,9 +21,16 @@ AI Zero Token provides a local CLI, web console, and HTTP gateway that expose sa
 - Account JSON import/export, including selected batch export.
 - Apply a saved account to local Codex by backing up and updating `~/.codex/auth.json`.
 - `gpt-image-2` image generation and JSON image editing through the ChatGPT internal Responses path.
+- Optional quota-exhaustion auto switch to the next saved API account with available quota.
 - Optional upstream proxy configuration for OAuth, model refresh, and gateway forwarding.
 - Local model discovery from the Codex model cache with manual refresh support.
 
+## Architecture
+
+![AI Zero Token architecture](docs/images/architecture-diagram.png)
+
+AI Zero Token keeps account tokens and gateway settings in local state, exposes OpenAI-compatible HTTP endpoints, and forwards requests to the selected ChatGPT/Codex OAuth account. The web console reads the same local state, so account switching, Codex auth application, proxy settings, and automatic switching are all controlled from one local place.
+
 ## Quick Start
 
 ```bash
@@ -40,6 +47,35 @@ http://127.0.0.1:8787/v1
 
 Use any non-empty API key value when a client requires one. Authentication is handled by the local gateway.
 
+## Desktop Preview
+
+This repository now includes an Electron desktop preview. It starts the existing local gateway from the desktop main process and loads the current web console:
+
+```bash
+npm run desktop
+```
+
+Build desktop release artifacts:
+
+```bash
+npm run dist:mac
+npm run dist:win
+```
+
+See [docs/DESKTOP_RELEASE.md](docs/DESKTOP_RELEASE.md) for release notes.
+
+The first desktop version keeps the current default listener:
+
+```text
+0.0.0.0:8787
+```
+
+For local clients, keep using:
+
+```text
+http://127.0.0.1:8787/v1
+```
+
 ## Web Console
 
 The web console is the recommended entry point:
@@ -50,6 +86,7 @@ The web console is the recommended entry point:
 - Export one account or selected accounts.
 - Apply a saved account to local Codex.
 - Configure the default text model and upstream proxy.
+- Enable automatic account switching when the active API account has exhausted its recorded quota.
 - Test `models`, `responses`, `chat.completions`, `images.generations`, and `images.edits`.
 
 ![AI Zero Token admin dashboard](docs/images/admin-dashboard.jpg)
@@ -187,6 +224,8 @@ The persistent state directory can be overridden with:
 AI_ZERO_TOKEN_HOME=/path/to/home azt start
 ```
 
+The web console settings are persisted in the same local state directory. The quota auto-switch option is stored as `autoSwitch.enabled`; when enabled, the gateway uses the latest saved quota snapshot and moves API traffic away from the active account once that snapshot shows a quota window is exhausted.
+
 The default request body limit is `32 MiB`, which is intended to make JSON base64 image references practical for local image editing. You can override it with:
 
 ```bash

package/README.zh-CN.md

@@ -21,9 +21,16 @@ AI Zero Token 是一个本地优先的 OpenAI 兼容网关，用于把 ChatGPT/C
 - 支持账号 JSON 导入/导出，以及复选框选择后的批量导出。
 - 支持把已保存账号应用到本机 Codex，并自动备份 `~/.codex/auth.json`。
 - 支持 `gpt-image-2` 文生图和 JSON 图生图。
+- 支持 API 账号额度耗尽后自动切换到下一个仍有额度的账号。
 - 支持上游代理配置，覆盖 OAuth、模型刷新和接口转发。
 - 模型列表优先读取本机 Codex 模型缓存，并支持手动刷新。
 
+## 技术架构
+
+![AI Zero Token 技术架构图](docs/images/architecture-diagram.png)
+
+AI Zero Token 会把账号 token、网关配置和运行状态保存在本地，通过 OpenAI 兼容接口转发到当前选中的 ChatGPT/Codex OAuth 账号。管理页读取同一份本地状态，因此账号切换、应用到 Codex、代理配置和额度耗尽自动切换都可以在本地统一管理。
+
 ## 快速开始
 
 ```bash
@@ -40,6 +47,35 @@ http://127.0.0.1:8787/v1
 
 如果客户端必须填写 API Key，可以填任意非空占位值；真正起作用的是本地网关里的账号授权。
 
+## 桌面端预览
+
+当前仓库已提供 Electron 桌面端预览入口。它会在桌面主进程中启动现有本地网关，并直接加载当前管理页：
+
+```bash
+npm run desktop
+```
+
+构建桌面端发布包：
+
+```bash
+npm run dist:mac
+npm run dist:win
+```
+
+发布说明见 [docs/DESKTOP_RELEASE.md](docs/DESKTOP_RELEASE.md)。
+
+桌面端第一版沿用现有默认监听策略：
+
+```text
+0.0.0.0:8787
+```
+
+本机客户端仍建议使用：
+
+```text
+http://127.0.0.1:8787/v1
+```
+
 ## 管理页
 
 管理页是推荐入口，可以完成：
@@ -50,6 +86,7 @@ http://127.0.0.1:8787/v1
 - 导出单个账号或勾选导出多个账号。
 - 将已保存账号应用到本机 Codex。
 - 配置默认文本模型和上游代理。
+- 开启当前 API 账号额度耗尽后的自动切换。
 - 测试 `models`、`responses`、`chat.completions`、`images.generations`、`images.edits`。
 
 ![AI Zero Token 管理页](docs/images/admin-dashboard.jpg)
@@ -187,6 +224,8 @@ AZT_CORS_ORIGIN=http://127.0.0.1:8124,http://localhost:3000 azt start
 AI_ZERO_TOKEN_HOME=/path/to/home azt start
 ```
 
+管理页里的配置会保存在同一个本地状态目录。额度耗尽自动切换会保存为 `autoSwitch.enabled`；开启后，网关会根据最近一次保存的额度快照判断当前 API 账号是否耗尽，并把 API 流量切到下一个仍有额度的账号。
+
 默认请求体上限是 `32 MiB`，用于让 JSON base64 图片在本地图片编辑场景里更实用。可以用下面的环境变量覆盖：
 
 ```bash