GCLI2API Admin panel

✨ Automation optimization: The system will now automatically enable required API services for your project after successful authentication

Gemini Cloud Assist API
Gemini for Google Cloud API

Description:No need to manually enable APIs, the system handles these configuration steps automatically for a smoother authentication process.

📁 Advanced options: Google Cloud Project ID (Ignore it, just click to get the link) ▶

Auth link:

Click this link to authenticate

Important:

Click the auth link above to open the Google OAuth page in a new window
Complete Google account sign-in and authorization
Redirects to localhost after successful authorization:11451Show success page
Close the OAuth window and return to this page
Click below"Get authentication file"Button Completes Flow

🚀 Cannot reach origin? Try the shortcut ▼

🚀 Antigravity Authentication mode

Get Google Antigravity credentials

Antigravity Auth link:

Click this link to authenticate

Instructions:

Click the auth link above to complete Google authorization in a new window
After authorization succeeds, it redirects to localhost showing a success page
Close the OAuth window and return to this page
Click below"Get credentials"Button Completes Flow

🚀 Cannot reach origin? Try the shortcut ▼

Batch Upload Authentication Files

Supports batch upload of GCLI and Antigravity auth files

📤 GCLI Bulk upload credentials

Click to select a file or drag it here

Supports .json and .zip files

ZIPThe file will be automatically decompressed to extract the JSON credentials

📤 Antigravity Bulk upload credentials

Click to select a file or drag it here

Supports .json and .zip files

ZIPThe file will be automatically decompressed to extract the JSON credentials

GCLICredential file management

Manage all GCLI credential files, view status and perform actions

📡 APIAddress (click to copy):

OpenAIFormat
ClaudeFormat
GeminiFormat

💡 Validation feature description:

Click each credential’s"Verify"Button can re-fetch Project ID, and view the account tier(free/pro/ultra)。
✅ Successful verification can resolve 403 errors，Make credentials work again.
Recommended when you hit a 403 error.

💡 Preview setting description:

Click each credential’s"Settings preview"Button can configure the Preview channel.
✅ Enabling preview can fix 404 errors when some credentials call the preview model。
Recommended when calling models like gemini-3-pro-preview and hitting 404 errors.

0 Total

0 Normal

0 Disable

Bulk actions

Select all

0 items selected

Loading credentials file...

AntigravityCredential file management

Manage all Antigravity credential files, view status and perform actions

📡 APIAddress (click to copy):

OpenAIFormat
ClaudeFormat
GeminiFormat

💡 Validation feature description:

0 Total

0 Normal

0 Disable

Bulk actions

Select all

0 items selected

Loading credentials file...

Configuration management

Manage system configuration parameters; changes take effect immediately

Loading configuration...

Server configuration

Server host address: Host address the server listens on; 0.0.0.0 means listen on all interfaces

Server port: Port the server listens on; restart the server after changing

APIAccess password: Chat API access password, used to authenticate OpenAI and Gemini API endpoints

Control Panel Password: Control panel access password, used for web interface login authentication

General password: （Kept for compatibility) Setting this overrides the two passwords above; leave blank to use the separate password settings

Basic configuration

Credential directory path: Directory path to store authentication files

Proxy settings: HTTP/HTTPS/SOCKS5Endpoint，Leave empty to not use a proxy

Endpoint configuration

Mirror URLs mainly solve the issue of not being able to reach official endpoints behind the firewall; may not work in some regions

Code Assist Endpoint: Google Cloud Code Assist APIEndpoint address

OAuth Endpoint: Google OAuth2 APIEndpoint address, used for token retrieval and refresh

Google APIs Endpoint: Google APIs APIEndpoint address for API service calls

Resource Manager API Endpoint: Google Cloud Resource Manager APIEndpoint URL for project management

Service Usage API Endpoint: Google Cloud Service Usage APIEndpoint address, used for service enablement management

Antigravity API Endpoint: Google Antigravity APIEndpoint address, used for anti-gravity mode

Auto-ban configuration

Enable auto ban Auto-disable credentials on specified error codes

Auto-ban error codes: Comma-separated list of error codes

Retry Configuration

Enable retry on error Auto-retry on error

Error retry count: Max retries on error

Error retry interval(seconds): Wait time between two retries when an error occurs

Compatibility Config

Enable compatibility mode When enabled, all system messages are converted to user; disables system_instructions ✓ Supports hot update

⚠️ Note:This option may reduce the model understanding ability, but can avoid empty streaming responses.
Use cases:Enable this option when the model returns no content or an empty response during streaming.

Return chain of thought to frontend When enabled, the model’s chain of thought is returned in the response; when disabled, it is filtered out ✓ Supports hot update

💭 Description:Some models (such as Gemini 2.0 Pro）Supports thinking mode, which outputs the reasoning process before generating a response. When enabled, you can see the model reasoning process; when disabled, only the final answer is shown for more concise output.

AntigravityStreaming to non-streaming When enabled, non-streaming requests use the streaming API and collect the full response ✓ Supports hot update

🔄 Description:Optimization option for Antigravity mode. When enabled, even if the client requests a non-streaming response, the backend uses the streaming API to fetch data and collects it fully before returning.
Use cases:In some cases the streaming API is more stable than the non-streaming API; enabling this option can improve response quality.
Default:Enabled

AntigravitySwitch credential When enabled, switch credentials on retry; when disabled, keep the current credential until it enters CD for the current model or is disabled ✓ Supports hot update

✅ Description:Off by default to allow continuous use of the same credential; still switches when a model cooldown or auto-disable occurs.

Anti-truncation settings

Max anti-truncation retries: Max continuation attempts when output truncation is detected

Note:The anti-truncation feature is now controlled by the model name:

Select with "Anti-truncation/" models with the prefix to enable
Both streaming and non-streaming requests work
e.g.: "Anti-truncation/gemini-2.5-pro"

Keep-alive configuration

Keep-alive URL: After configuration, the service periodically sends GET requests to this URL to stay online; leave blank to disable keep-alive ✓ Supports hot update

Keep-alive interval (seconds): Wait time between keep-alive requests, range 5 - 86400 seconds, default 60 ✓ Supports hot update

Hot reload configuration notes

🔥 Hot reload config (takes effect immediately):

Network configuration:Proxy settings, endpoint configuration, HTTP timeout, max connections
APIConfiguration:Credential rotation count, error retry settings, auto-ban configuration
Password configuration:APIPassword, control panel password, general password
Feature configuration:Max anti-truncation retries

🔄 Settings requiring restart:

Server configuration:Host address, port number
Directory configuration:Credential directory path, Code Assist endpoint

Live logs

View real-time system log output, with log filtering and auto-scroll

Connection status: Not connected

Waiting to connect to log stream...

Project info

Details and support for the GCLI2API project

📋 Project description

GCLI2APIA proxy that converts Google Gemini API to OpenAI and GEMINI API formats, supporting multi-account management, automatic rotation, real-time log monitoring, and more.

🔗 Project address: GitHub - su-kaka/gcli2api

⚠️ Usage statement: No commercial use or resale - for learning only

✨ Main features

🔄 Multi-account management: Supports batch upload and management of multiple Google accounts

⚡ Auto-rotate: Smart account rotation avoids single-account limits

📊 Real-time monitoring: Usage statistics, error monitoring, real-time logs

🛡️ Safe and reliable: OAuth2Authentication, automatically ban abnormal accounts

🎛️ Flexible configuration: Supports hot reload config, proxy settings

📱 User-friendly interface: Responsive design, mobile support

💬 Community Group

Welcome to join the QQ group for discussion!

QQ Group ID: 1083250744

Scan to join QQ group

📞 Contact us

• Feedback: Submit issues and suggestions via GitHub Issues

• Feature request: Discuss new features in GitHub Discussions

• Code contributions: Welcome to submit Pull Requests to improve the project

• Documentation improvements: Help improve project docs and usage guide

GitHub: https://github.com/su-kaka/gcli2api

⚠️ No commercial use or resale - for learning only ⚠️

GCLI2API Admin panel

Auth link:

Authentication file content:

Antigravity Auth link:

Antigravity Credential content:

Batch Upload Authentication Files

📤 GCLI Bulk upload credentials

Selected files:

Upload progress:

📤 Antigravity Bulk upload credentials

Selected files:

Upload progress:

GCLICredential file management

Bulk actions

AntigravityCredential file management

Bulk actions

Configuration management

Server configuration

Basic configuration

Endpoint configuration

Auto-ban configuration

Retry Configuration

Compatibility Config

Anti-truncation settings

Keep-alive configuration

Hot reload configuration notes

Live logs

Project info

📋 Project description

✨ Main features

💬 Community Group

📞 Contact us