Documentation

InferenceHub is a byte-faithful gateway: one API key for Claude, GPT, and every model, served on their native APIs with 0% per-token markup. Point any compatible client at the gateway and keep your existing code.

Using an AI agent? Read these docs as raw markdown or point it at our llms.txt.

How it works

Create a key in the portal — every account starts with $1 in free credit, no card required.
Point your client at the gateway — it speaks the Anthropic and OpenAI wire formats your tools already use.
Pick any model with the usual model field — one key reaches the whole catalog at native per-token rates.

Three minutes end to end — the quickstart has copy-paste requests for both wire formats.

Connect your coding agent

Each agent has a dedicated page with the exact config, which models it reaches, and its known gotchas:

Claude Code

Claude and the open-weight daily drivers in Claude Code.

Codex

OpenAI Codex on the Responses wire.

Cursor

Open-weight, Claude, and GPT-5.6 as Cursor custom models.

Kilo Code

The full catalog from one kilo.json — CLI and VS Code extension.

ZCode

Z.ai's GLM-5.2 harness on your InferenceHub key.

Hermes

Open-weight chat models in the Hermes CLI.

Explore

API overview

Wire formats, base URLs, authentication, streaming.

Image generation

OpenAI Images API — gpt-image-2 and gpt-image-1.5.

Embeddings & rerank

Embedding and reranker models for RAG pipelines.

Models

Model ids, aliases, and how to list what your key reaches.

Auto routing

ih-auto and ih-auto-smart — a flat sheet, per-request routing, and honest escalation.

Model synthesis

ih-synthesis presets and the model_synthesis tool — a model panel answers, a synthesizer merges.

Billing & plans

Prepaid balance, subscription plans, free chat models, and the daily reward.

Desktop app

The open-source native client for the hosted chat.

Troubleshooting

Common HTTP errors and per-agent fixes.

Support

Discord, Telegram, email, and the status page.