APIYI - AI Model API Hub

Enterprise-grade AI API aggregation: one key for 400+ models, compatible with OpenAI, Claude, and Gemini native formats, operating for 2+ years with 150K monthly visitors

中文用户？ 切换到中文版

APIYI is an enterprise-grade, professional and stable AI model API hub — compatible with the unified OpenAI API standard while also supporting each vendor’s native API formats (Claude, Gemini, and more), covering 400+ popular AI models. With one token, you can easily access OpenAI, Claude, Gemini, DeepSeek, Qwen, Kimi, GLM, Minimax and all mainstream large language models.

🎬 Start Here: APIYI in 40 Seconds

One key for 270+ models, switch models by changing one line, official routes with no downgrades — and your first API call in three minutes.

🏢 Company Background

• Operating Entity: APIYI, LLC (United States)
• Official Partners: Google AI Studio, Microsoft Azure, Amazon AWS (legitimate quota sources, use with confidence)
• Service Guarantee:
  • Stable: Provides high concurrency and stable services for mainstream models like OpenAI, Claude, Google Gemini
  • Trusted:
    • Many well-known applications have stably integrated in production environments (see Use Cases section).
    • Cooperation with renowned universities, hospitals and other institutions, serving well-known enterprises.
    • Domestic entities can make corporate payments, issue invoices, and assist with procurement lists for worry-free reimbursement.

🛡️ Tired of Sketchy API Resellers?

The API reseller market is a mixed bag — silently swapped models, degraded quality, and vanishing operators are all too common. APIYI doesn’t rely on verbal promises; we give you verifiable evidence.

2+ Years in Operation

150K monthly visitors, long-term stable operation — not a fly-by-night newcomer

30-Day No-Questions Refund

Unused balance refunded to the original payment method within 30 days, no fees via WeChat/Alipay

Pure Official Relay · Model Fidelity

No rerouting, no degradation, no model swapping — verify fidelity yourself

Pragmatic SLA Guarantee

Compensation for billing anomalies, credit reissue for outages, contractual SLA for enterprises

Data Security

Transparent proxy, no conversation content retained, minimal logging

Corporate Payment · Easy Reimbursement

Invoices and corporate transfers supported — reimbursement-friendly for universities, hospitals and enterprises

Where APIYI Fits

Reliability and peace of mind come first. But let’s be honest: quality resources aren’t cheap, and cheap ones always come at another cost — swapped models, degraded output, throttling, or an operator that vanishes, and your business pays for it in the end.

Use APIYI as Your Primary

Get a high-quality, stable, full-fidelity API experience and keep your focus on your product

Keep APIYI as Your Backup

When your main provider has trouble, we’ve got you covered so your business keeps running

🤖 Three ways to let an AI integrate for you

No reading docs, no filling in config by hand. Pick whichever AI you have at hand, copy the matching prompt to it, and it does the rest.

Skills · chat agents

Call APIYI in natural language from a chat agent such as OpenClaw. Copy the prompt to the agent; it installs the skill, self-checks the key and walks you through setup.

Copy to OpenClaw, Claude Code, Cursor or any agent that supports skills

CopiedCopy prompt

Skill source github.com/apiyi-com/skills · install routes and key setup in the AI Developer Kit

CLI · terminal

Get the first call working from the terminal without writing a line of code. npx apiyi@latest check needs no install: it checks the key, measures node latency, lists models, sends messages and generates images.

Copy to any agent that can run terminal commands, or run the lines yourself

CopiedCopy prompt

Source github.com/apiyi-com/cli · npm package apiyi · full command table in the AI Developer Kit

AI Developer Kit · coding agents

Make Cursor / Claude Code / Codex read the contract before writing code. skill.md is the rulebook, llms.txt the index, model-registry.json the source of truth for models. Read, explain, then edit. No invented endpoints.

Copy to Cursor, Claude Code, Codex or any other coding agent

CopiedCopy prompt

What is in the kit and how each file is used: AI Developer Kit

🌟 Featured Models

Nano Banana Pro/2 Series

🎨 Image Generation

The best available, 4K for just

$0.09 /image

🎯 As low as 30.3% of official price

Claude Official Relay

🤖 Chat & Coding

Opus 5.5 / Sonnet 5.5 / Haiku 5.5, pure official relay via AWS + Official dual channels

⚡ ~80% of official price

GPT-image-2 Full Series

🖼️ Image Generation

Official relay + reverse (-all/-vip) full coverage

🎯 Pick what fits your needs

Gemini Full Multimodal Series

🔥 Multimodal

Full lineup for text, image, and video

🏆 Gemini 4 Argon flagship now live

OpenAI Full Multimodal Series

🚀 Multimodal

GPT-6.1 Sol / GPT-6 series, image, and audio

✅ Full lineup, stable support

Full Video Generation Lineup

🎬 Video Generation

Seedance 2.5 / Wan3.0 / MiniMax H3 / Oxygen / VEO 3.1

🆕 Per-second billing from $0.02

📖 Product Basics

Quick Start

Complete integration in three steps, start using AI models immediately

API Manual

Complete API documentation and developer guide

AI Developer Kit

🤖 Skill, CLI, and the contract plus model registry for coding agents

Changelog

🔥 Latest model launches, price adjustments and important updates

Pricing

View detailed pricing and promotions for all models

🔧 Core APIs

Chat Completions API

Chat Completions - Create multi-turn conversations and text generation

Models API

Models API - Get all available model information

Image Generation API

Images API - Nano Banana Pro, gpt-image-2 and more

Embeddings API

Embeddings API - Text vectorization and semantic search

⚡ API Capabilities

🎬 Video Generation API

Seedance 2.5 / 2.0 Video Generation

🔥 ByteDance’s latest flagship: 2.5 renders 30-second clips with up to 30 reference images; 2.0 offers standard / fast / mini tiers, synchronized audio by default

Wan3.0 Video Generation

🆕 One model ID for text, image, and reference-to-video plus video editing, up to 30 seconds with audio by default, from $0.0294/second

MiniMax H3 Video Generation

🆕 Omni-modal video model with mixed image / video / audio references and built-in stereo, $0.03/second

Oxygen Video Generation

🆕 OpenAI Videos-compatible API: text, first/last-frame, and reference-to-video in one model, $0.02/second

VEO 3.1 Video Generation

Google official channel, per-generation billing from $0.3/video, 720p / 1080p / 4K

Character-Consistent Video

Seedance asset library: upload images once, reference them for character-consistent videos

Video Understanding API

Intelligent video analysis, scene recognition, content understanding

🎨 Image Generation API

Nano Banana Pro

🔥 Our strongest, 4K HD, best-in-class text rendering, $0.09/image

Nano Banana 2.1

🆕 The Nano Banana 2 upgrade with better image quality and text rendering, $0.045/image per request, same price for 1K–4K

Nano Banana 2

🔥 Usage-based billing, new 1:8/8:1 long-image ratios, $0.055/image (from $0.025 usage-based)

Nano Banana Lite

🆕 Google’s fastest and cheapest, ~4s per image, $0.025/image

gpt-image-2.5 / 2 Series

OpenAI official relay 2.5-flare / 2.5-sunburst / 2 with native 4K + reverse -all at $0.025/image and -vip at $0.03/image — pick what fits

Seedream 5.0 Pro / Flash

BytePlus official partnership, Flash renders in 15–20 seconds from $0.018/image, URL output

MAI-Image 2.6

🆕 Microsoft’s in-house image model via official relay, excellent Chinese text rendering, Flash from $0.06/image

Flux 2 Series

🆕 FLUX.2 max/pro/flex tiers, from $0.03/generation

🔧 Basic APIs

Model Info

View detailed information for 400+ supported AI models

OpenAI SDK Integration

Seamlessly integrate APIYI using the official SDK

Image Understanding API

Intelligent image analysis, OCR, object recognition, scene description

Claude Native Format

Anthropic native API format, full feature support

Gemini Native Format

Google native API format, full feature support

Text Embedding API

Text vectorization and semantic search

🎯 Use Cases

💬 Conversational AI

Cherry Studio

Powerful AI chat client with multi-model switching

Chatbox

Cross-platform desktop AI chat application

Open WebUI

Self-hosted web chat interface

ChatGPT Next Web

One-click deployable web-based ChatGPT

💻 Programming & Development

Claude Code

🔥 Anthropic’s official AI programming assistant

Cursor

AI-powered code editor

Cline (VS Code)

AI programming assistant in VS Code

Roo Code

Efficient AI code generation tool

Codex CLI

Command-line AI programming assistant

Gemini CLI

Google Gemini command-line tool

🔧 Engineering

LangChain

Development framework for building AI applications

Dify

Visual AI application development platform

🌐 Translation

Bob Translator

Professional translation tool for macOS

Immersive Translate

Browser extension for bilingual reading

🚀 Why Choose APIYI?

One Interface, Multiple Models

No need to apply for separate accounts and manage API keys for each AI service. With APIYI, you only need:

• One account: Manage all AI services
• One API key: Access all models
• One standard: Compatible with OpenAI API format

💡 Supported Models

We support 400+ industry-leading AI models:

OpenAI Series

• GPT-6.1 Sol (🔥 Newest, intelligence index just 1 point below Astra, cache reads halved)
• GPT-6 Astra (6-gen flagship for computer use and long-horizon agents)
• GPT-6 Sol / GPT-6 Luna (about half the price of 5.6, long-term pricing)
• GPT-5.6 Sol / Terra / Luna, GPT-5.5 / GPT-5.4
• GPT-5.5 Pro (deep reasoning, Responses endpoint only)
• gpt-image-2.5 Flare / Sunburst, gpt-image-2 (image generation)
• gpt-realtime-2.1 / 2.1-mini (realtime voice)

Anthropic Series

• Claude Opus 5.5 (🔥 Flagship, cheaper than Opus 5)
• Claude Sonnet 5.5 (🔥 Terminal-Bench 4.0 70.6%, same price as Sonnet 5)
• Claude Haiku 5.5 (🔥 1M context, $0.10 input / $0.50 output per million within 100K tokens)
• Claude Fable 5.1 (Mythos-class flagship)
• Claude Opus 5 / Claude Sonnet 5
• Claude Opus 4.8 / Opus 4.7 / Sonnet 4.6 / Haiku 4.5

Google Series

• Gemini 4 Argon (🔥 First Gemini 4 flagship, same accuracy with far less thinking)
• Gemini 3.8 Flash (Latest Flash, same price as 3.7)
• Gemini 3.7 Flash / 3.6 Flash / 3.5 Flash-Lite
• Gemini 3.1 Pro Preview
• Nano Banana 2.1 (🔥 $0.045 per request, same price for 1K–4K), Nano Banana Pro / Nano Banana 2

xAI Grok Series

• Grok 4.7 (🔥 Newest flagship on a 2.1T base)
• Grok 4.6 / Grok 4.5 (code & agents)
• Grok 4.3 (1M context)
• API guide (web search / X search / code execution verified)

Chinese Models

• DeepSeek V4.1 Flash / V4 Pro
• GLM-5.3 / GLM-5.3-Flash (Zhipu AI)
• Kimi K3 (Moonshot)
• Qwen3.8-Max / Qwen3.7 series (Alibaba)
• MiniMax-M3
• Seed 2.1 Turbo (ByteDance)

Video Generation Models

• Seedance 2.5 (🔥 ByteDance’s latest, 30-second clips, up to 30 reference images)
• Seedance 2.0 series (standard / fast / mini)
• Wan3.0 (🔥 Alibaba Wan’s new generation, standard / prime tiers), Wan2.7, HappyHorse
• MiniMax H3 (omni-modal references, built-in stereo)
• Oxygen 1.0 (built by AZ8, $0.02/second for high-volume work)
• VEO 3.1 (Google official channel, up to 4K)

Image Generation Models

• gpt-image-2.5 (🔥 Flare for speed / Sunburst for quality)
• Nano Banana 2.1 (🔥 same per-request price for 1K–4K), Nano Banana Pro / Nano Banana 2
• Seedream 5.0 Flash (🔥 $0.018/image), Seedream 5.0 Pro
• MAI-Image 2.6 / 2.6 Flash (Microsoft in-house, excellent Chinese text rendering)

🔧 Simple & Easy

Switching models is as simple as changing one parameter:

from openai import OpenAI

client = OpenAI(api_key="sk-your-apiyi-key", base_url="https://api.apiyi.com/v1")

# Using GPT-6.1 Sol
response = client.chat.completions.create(
    model="gpt-6.1-sol",
    messages=[{"role": "user", "content": "Hello!"}]
)

# Switch to Claude Opus 5.5
response = client.chat.completions.create(
    model="claude-opus-5-5",  # Just change the model name
    messages=[{"role": "user", "content": "Hello!"}]
)

🛡️ Stable & Reliable

• High Availability: Multi-node deployment, intelligent routing
• Auto Failover: Automatic switching when a model is unavailable
• Load Balancing: Intelligent request distribution, avoiding rate limits
• Real-time Monitoring: 24/7 service status monitoring

💰 Cost Optimization

• Unified Billing: All models use a unified balance
• Transparent Pricing: Clear pricing structure
• Usage Statistics: Detailed usage reports
• Flexible Top-up: Multiple payment methods supported

🎯 Key Features

🔥 Latest Models Available Immediately

• Gemini 4 Argon (live on day one, 14 October): matched 3.1 Pro at 36/36 in our tests with about 1/12 of the thinking tokens
• Claude Opus 5.5 / Sonnet 5.5 / Haiku 5.5: Sonnet 5.5 scores 70.6% on Terminal-Bench 4.0; Haiku 5.5 supports 1M context
• GPT-6.1 Sol: intelligence index just 1 point below GPT-6 Astra, cache reads down to $0.10 per million tokens
• Wan3.0 Video Generation: one model ID for text, image, and reference-to-video plus video editing, up to 30 seconds

🚀 Stable, Reliable & Unlimited Concurrency

Official partner resources (AWS, Azure, Google Cloud, BytePlus), high-performance infrastructure support, unlimited concurrency, ensuring stable operation in multi-industry production environments.

💰 Ultimate Value

• Top-up bonus: Up to 80% discount
• Exchange rate advantage: USD pricing is more affordable
• Cache optimization: cached-token billing on OpenAI, Claude, and Gemini cuts the cost of repeated context
• Pay-as-you-go: Flexible billing by token or per use

🚀 Get Started

Ready to begin? Just three steps:

Register

Create your APIYI account

Get API Key

Generate your API key

Integration

View API docs to start integration

🔗 Quick Links

Register Now

New accounts come with $0.05 in trial credit — make your first call without topping up

Dashboard

Manage API Keys, view usage statistics and billing

---

New accounts come with $0.05 in trial credit — enough to run a Hello World against a lightweight model such as gpt-5.4-mini and confirm your integration works. Top up when you are ready for real usage; see the recharge bonus policy.