AIone API (English)
    • 01 - Quick Start
    • 02 - Authentication
    • 03 - Error Codes
    • 04 - Pricing
    • 05 - Contact Us
    • 06 - Quality of Service
    • 07 - Complete Examples
    • 08 - Caching & Cost Optimization
    • 11 - Model Quality Monitoring
    • 12 - Network & Connectivity
    • 13 - Model Naming & Compatibility
    • 14 - Gemini Image Generation
    • 09 - Model Verification
    • 10 - IDE Integration

    01 - Quick Start

    Quick Start#

    1. Get Your API Key#

    Log in to the AIone Console and create a new key on the API Keys page.
    We recommend creating separate keys for each environment or use case to simplify access control, usage tracking, and troubleshooting.

    2. Send Your First Request#

    AIone is fully compatible with the OpenAI Chat Completions API. Simply set base_url to https://api.nexara.net/v1 and use the OpenAI SDK or any compatible client.
    Streaming is recommended: For interactive use cases, always enable stream: true. This reduces time-to-first-token from 10+ seconds down to 2-3 seconds, significantly improving the user experience.

    Python (Streaming - Recommended)#

    Python (Non-Streaming)#

    Node.js (Streaming - Recommended)#

    cURL (Streaming)#

    3. Switch Models#

    All models are served through the unified v1/chat/completions endpoint. Simply change the model parameter to switch:
    Modelmodel ParameterBest For
    Claude Sonnet 4.6claude-sonnet-4-6General-purpose; balanced speed and quality
    Claude Opus 4.6claude-opus-4-6Strongest reasoning; complex tasks
    Claude Sonnet 4.5claude-sonnet-4-5-20250929High-quality general reasoning and code
    Claude Opus 4.5claude-opus-4-5-20251101Top-tier complex analysis
    Claude Haiku 4.5claude-haiku-4-5-20251001Ultra-fast responses; high concurrency
    Claude Sonnet 4claude-sonnet-4-20250514Stable release; broad compatibility
    GPT-5.4gpt-5.4OpenAI's current flagship
    GPT-5.2gpt-5.2High-quality general and coding tasks
    GPT-5gpt-5Existing integration compatibility
    GPT-5 Minigpt-5-miniHigh concurrency; low latency
    GPT-4.1gpt-4.1Code and complex instruction following
    GPT-4ogpt-4oMultimodal and general tasks
    Gemini 2.5 Progemini-2.5-proComplex reasoning and long context
    Gemini 2.5 Flashgemini-2.5-flashSpeed-first scenarios
    Gemini 2.5 Flash Litegemini-2.5-flash-liteLower cost; high concurrency
    For the full model list, visit the Models & Pricing page in the console or query the /v1/models endpoint.

    4. Try the Sandbox#

    After logging in, go to the Sandbox page to test any model directly in your browser -- no code required.
    Modified at 2026-04-04 16:06:55
    Next
    02 - Authentication
    Built with