100+ Full-Capability Official Models

Unified AI API Gateway forClaude, GPT and Gemini

Access 100+ models through a single unified endpoint, fully compatible with OpenAI, Anthropic and Gemini protocols — change one line to unlock them all.

Get API Key FreeBrowse Models
100+Models
5+Providers
3Protocols
95%+Cost Saving
NEWClaude MAX group · native Anthropic access for Claude Code

Popular Models

Transparent pricing, pay only for what you use.

Hot

claude-opus-4-8

Anthropic
Available
Input
$0.75 /1M
Output
$3.75 /1M

claude-opus-4-7

Anthropic
Available
Input
$0.75 /1M
Output
$3.75 /1M

claude-sonnet-5

Anthropic
Available
Input
$0.15 /1M
Output
$0.75 /1M

gemini-3.5-flash-high

Google
Available
Input
$0.01 /1M
Output
$0.06 /1M

gpt-5.5

OpenAI
Available
Input
$0.06 /1M
Output
$0.24 /1M

gpt-5.4

OpenAI
Available
Input
$0.06 /1M
Output
$0.24 /1M
View All Models
Use it anywhere

Runs right inside the tools you already use

Web console, command line, desktop apps and mobile — one API key, everywhere.

https://zivv.pro/dashboard
Balance
$248.50
Live
Requests
1.2M
Tokens
845M
Models
17

// web console

claude-code — zsh
$ claude --model sonnet
→ connected · zivv gateway
✓ Ready — start chatting
$

// CLI / Claude Code

// mobile

Any client that supports a custom endpoint. 30-second setup.

Why Zivv API?

Everything you need for production-grade AI integration.

Unified Interface

Single endpoint for OpenAI, Anthropic, and Google. Fully compatible with OpenAI, Anthropic and Gemini protocols — zero migration cost.

Smart Routing

Automatic failover and load balancing across channels. Your requests always find the fastest available path.

Cost Efficient

Unified billing with significant savings compared to direct provider pricing. No hidden fees.

Real-time Monitoring

Track every token, monitor usage per model and per key. Full visibility into your API consumption.

Team Collaboration

Manage shared balance, member permissions, individual keys and budgets, with clear usage by member and project.

Stable Performance

Multi-route scheduling, automatic failover and load balancing keep high-concurrency and long-running work available.

Enterprise Security

Requests are encrypted in transit, request and response content is not logged, and keys are encrypted with access isolation.

Fully In-House

The entire product and codebase are built in-house — not a rebranded open-source gateway. Faster fixes, independent iteration, and long-term stability.

Price Advantage

Access top-tier AI models at a fraction of official API pricing through Zivv API.

Claude

Anthropic

5%

of official API price

cost saved 95%

Anthropic Native

Built for Claude

  • Full Anthropic Messages API support — not just OpenAI translation
  • Claude Code, Cursor, Windsurf direct connection
  • Extended thinking, tool use, vision & all advanced features
  • Also fully compatible with OpenAI and Gemini protocols for all other models

GPT

OpenAI

3%

of official API price

cost saved 97%

OpenAI Compatible

OpenAI Compatible

  • Fully compatible with OpenAI Chat Completions API
  • All models accessible via OpenAI SDK directly
  • Streaming, function calling, vision & all features supported
  • Zero migration — just change your base URL

Gemini

Google

5%

of official API price

cost saved 95%

Gemini Native

Gemini Native Protocol

  • Full Gemini generateContent API support — native protocol, not translation
  • All Gemini models supported — new models added on day one
  • Native multimodal, function calling, grounding & all advanced features
  • Also fully compatible with OpenAI and Anthropic protocols

Quick Start

Select protocol, copy code, start calling

BASE URLhttps://your-domain.com/v1
from openai import OpenAI

client = OpenAI(
    api_key="your API Key",
    base_url="https://your-domain.com/v1",
)

response = client.chat.completions.create(
    model="claude-sonnet-4-6",
    messages=[{"role": "user", "content": "Hello"}],
)
print(response.choices[0].message.content)

Integration Tools

Claude CodeCherry StudioOpenClawMore Integrations

Frequently Asked Questions

Zivv API is a unified AI gateway that lets you access GPT, Claude, Gemini and other models through a single API endpoint. Fully compatible with OpenAI, Anthropic and Gemini protocols.
Pay-as-you-go based on actual token usage. Each model has transparent per-million-token pricing. No monthly minimums or hidden fees.
All major models from OpenAI, Anthropic, and Google are supported, including the latest flagship and lightweight models. New models are added as soon as they launch. See the full list on the Model Hub.
All requests are encrypted in transit. We do not store or log any request/response content. Your API keys are encrypted at rest.
Nearly all popular AI coding tools: Claude Code, OpenAI Codex, OpenCode, OpenClaw, Cursor, VS Code (Continue/Cline), Windsurf, CherryStudio and more. Any tool that supports custom API endpoints can connect to Zivv. Setup takes just 30 seconds.
Charges are calculated as 'group multiplier × model official price'. Check exact pricing per model on the Model Hub. Billed by actual token usage with no hidden fees. Top-up rate: ¥1 = $1.

Start building in 5 minutes

Get your API key, change one line of code, and access all models instantly.

Create Free AccountView Documentation