Nexflow Welcome to Nexflow
Nexflow is a decentralized AI inference aggregator designed to route developer prompts across frontier models with sub-second latency, powered by the $NF token economy.
The Core Philosophy
We believe builders shouldn't be locked into a single AI provider. With one unified API endpoint, your applications instantly gain multi-model redundancy across OpenAI, Anthropic, Gemini, Grok, and Llama.
Getting Started in 30 Seconds
Generate your API key from the dashboard, choose your model routing preference, and start shipping immediately.
Changelog
Stay updated with the latest improvements, protocol upgrades, and new model integrations.
Smart Failover & Staking Live
We're thrilled to introduce Smart Failover. Nexflow now intelligently monitors provider latency in real-time. If your primary provider (e.g., GPT-4o) hits a bottleneck, we instantly switch your request to the next best model (e.g., Claude 3.5 Sonnet) without you writing a single line of extra code.
Staking is officially live for $NF. Node operators can now lock their tokens to participate in decentralized inference verification.
Grok-2 & Llama 3.1 Support
Expanded our model library. You can now access Grok-2 and Llama 3.1 405B via the same standard endpoint. We've optimized throughput lanes for these models to ensure 18ms latency or less.
Mainnet Launch
Nexflow Mainnet is live on Arc. Settlement for all inference requests is now gasless and sub-second. $NF utility enabled.
System Status
Nexflow monitors every endpoint globally. We aim for 99.99% uptime for your production AI workloads.
API Reference
Nexflow uses a RESTful API standard that is fully compatible with the OpenAI SDK. If you can build with OpenAI, you can build with Nexflow.
Authentication
All requests must include your API key in the Authorization header.
Authorization: Bearer nx_live_YOUR_KEY_HERE
Chat Completions
To send a message, make a POST request to our chat endpoint:
POST https://api.nexflow.ai/v1/chat/completions
Nexflow will automatically route your request based on the model parameter provided. Use nexflow-auto to let us handle the best model selection for your specific task.
Integrations & SDKs
Integrate Nexflow in minutes. We provide officially supported SDKs and community-driven wrappers to make your development journey seamless.
Nexflow Python SDK
Our official Python library simplifies complex routing and retries.
pip install nexflow-ai
View on GitHub →
LangChain Integration
Use NexflowChat as your primary LLM interface within LangChain.
from langchain_nexflow import NexflowChat
$NF Tokenomics & Utility
The $NF token powers governance, decentralized inference routing, and node staking on Arc.
Token Details
Ticker: $NF
Network: Arc
Mechanism: 10% Burn on API inference spend + Staking rewards.