Skip to main content
NexLLM is a unified AI model gateway that lets you call GPT, Claude, and Gemini models through one consistent, OpenAI-compatible API endpoint. Instead of juggling separate SDKs, credentials, and billing accounts for every provider, you point your existing OpenAI client at https://www.nexllm.ai/v1, swap in your NexLLM API key, and start making requests immediately — no other code changes required.

What you can do with NexLLM

  • Access multiple model families from a single endpoint. Send requests to OpenAI, Anthropic, and Google models using the same base URL and authentication scheme.
  • Organise capacity with channel groups. Channel groups let you route tokens to different pricing tiers or upstream configurations, giving you granular control over cost and throughput across teams or projects.
  • Test models without writing code. The built-in Playground lets you chat with any available model directly in your browser, making it easy to compare outputs or verify a token before deploying.
  • Monitor usage in real time. The usage dashboard surfaces per-key token consumption, request counts, and latency logs so you can track spend and debug issues without leaving the platform.
  • Pay through a wallet balance. NexLLM uses wallet-based billing — top up your balance and usage is deducted automatically, giving you a single invoice across all providers.

Base URL

All API requests are sent to:
https://www.nexllm.ai/v1
This endpoint is fully compatible with the OpenAI API schema, so any library that accepts a configurable base_url works out of the box.

Explore the docs

Registration

Create your NexLLM account and sign in for the first time.

API Keys

Generate, configure, and secure your API keys.

Quickstart

Make your first API call in minutes using curl or the Python SDK.

API Reference

Explore the full list of supported endpoints and request schemas.

Channel Groups

Learn how to route requests across pricing tiers and upstream providers.

Models

Browse all available GPT, Claude, and Gemini models and their identifiers.