Misc Public API

3. Group List (Group)

3.1 Get All Group Names

Endpoint: GET /api/group/

Auth: Admin

Request parameters: None

Response:

{
  "success": true,
  "message": "",
  "data": ["default", "vip", "svip"]
}

Description: Returns every group name defined in the group_prices table.

4. System Options (Option)

4.1 Get All System Options

Endpoint: GET /api/option/

Auth: Root

Request parameters: None

Response:

{
  "success": true,
  "message": "",
  "data": [
    { "key": "QuotaForNewUser", "value": "1000000" },
    { "key": "QuotaPerUnit", "value": "500000" },
    { "key": "TopUpLink", "value": "" },
    ...
  ]
}

Description: Sensitive options (tokens, SMTP passwords, etc.) are filtered or masked.

4.2 Update a System Option

Endpoint: PUT /api/option/

Auth: Root

Request body:

{
  "key": "QuotaPerUnit",
  "value": "500000"
}

Request fields:

Field	Type	Required	Description
key	string	yes	Option name
value	string	yes	Option value

Available options:

Option name	Description
QuotaForNewUser	Quota granted to new users
QuotaForInviter	Reward quota for the inviter
QuotaForInvitee	Reward quota for the invitee
QuotaRemindThreshold	Quota reminder threshold
PreConsumedQuota	Pre-consumed quota
TopUpLink	Top-up link
ChatLink	Chat link
QuotaPerUnit	Quota per unit (how much internal quota equals ¥1)
DisplayInCurrencyEnabled	Whether to display amounts as currency (true/false)
DisplayTokenStatEnabled	Whether to display token stats (true/false)
ApproximateTokenEnabled	Whether to use approximate token counting (true/false)
RetryTimes	Retry count
LogConsumeEnabled	Whether to enable consumption logs (true/false)
AutomaticDisableChannelEnabled	Auto-disable channels (true/false)
AutomaticEnableChannelEnabled	Auto-enable channels (true/false)
ChannelDisableThreshold	Channel disable threshold
ChannelDefaultCooldownSeconds	Default cooldown (seconds)
ChannelMaxCooldownSeconds	Max cooldown (seconds)
ChannelConcurrencyEnabled	Enable per-channel concurrency limit (true/false)
ChannelStickySessionEnabled	Enable sticky session (true/false)
ErrorNext	Error response strategy JSON

Response:

{
  "success": true,
  "message": ""
}

14. Other Public Endpoints

Endpoint	Method	Description
/api/status	GET	Get system status (no auth required)
/api/notice	GET	Get system notice
/api/about	GET	Get about page content
/api/home_page_content	GET	Get home page content
/api/verification	GET	Send email verification code
/api/reset_password	GET	Send password reset email
/api/user/reset	POST	Reset password
/api/oauth/github	GET	GitHub OAuth callback
/api/oauth/oidc	GET	OIDC OAuth callback
/api/oauth/lark	GET	Lark OAuth callback
/api/oauth/state	GET	Generate OAuth state code
/api/oauth/wechat	GET	WeChat OAuth callback
/api/oauth/wechat/bind	GET	Bind WeChat account (requires login)
/api/oauth/email/bind	GET	Bind email (requires login)
/api/user/logout	GET	Logout
/api/user/aff	GET	Get affiliate code
/api/user/subscription	GET	Get subscription info

Appendix B: Permission Levels

Level	Value	Description
Guest	0	Unauthenticated visitor
User	1	Regular user
Admin	10	Admin
Root	100	Super admin

Appendix C: Billing Types

billing_type	Description
token	Per-token billing, priced in ¥ per million tokens
per_request	Per-request billing, priced in ¥ per request

Token billing formula:

quota = ceil((inputPrice × inputTokens + outputPrice × completionTokens + cachedPrice × cachedTokens) / 1,000,000 × groupDiscount × QuotaPerUnit)

Per-request billing formula:

quota = ceil(perRequestPrice × sizeRatio × N × groupDiscount × QuotaPerUnit)

Group discount matching rules:

1. GroupName + ModelName exact match → use that discount.
2. GroupName + "" (empty) → use that group's default discount.
3. No match → discount is 1.0 (no discount).

QuotaPerUnit: Default 500,000, meaning 500,000 internal quota equals ¥1.

Appendix D: Channel Type Reference

Type	Adaptor
1	OpenAI
2	Azure OpenAI
3	Custom channel
4	Claude (Anthropic)
5	Google Gemini
6	Tongyi Qianwen (Ali)
7	iFlytek Spark (SparkDesk)
8	Baidu ERNIE (Baidu)
9	ByteDance Doubao (Doubao)
10	MiniMax
11	DeepSeek
12	Cohere
13	360 Brain
14	Ollama
15	Moonshot
16	Zhipu AI (GLM)
17	Baichuan
18	Yi (01.AI)
19	DeepSeek (dedicated adaptor)
20-40+	Other adaptors