Pay as you go
Top up, then spend per token
Usage basedno monthly fee
- Billed from the list price of every model
- Card, wallet and crypto top-ups
- Per-request detail in the console
- Balance works across all 400+ models
One API key for 400+ large language models
QuickRouter.AI - the LLM API platform for builders
Direct, high-speed access to OpenAI, Claude, Gemini, DeepSeek, Grok and 400+ more models. Pay for what you use, skip the monthly fee, and keep the OpenAI SDK you already wrote.
- One API key, every model
- OpenAI SDK compatible
- No overseas card required
- 400+
- models behind one endpoint
- 99.9%
- gateway uptime target
- 3 min
- from signup to first call
- 1
- base URL to remember
OpenAIAnthropicGoogleDeepSeekxAIQwenMetaMistralZhipuMoonshotMiniMaxByteDanceKlingViduOpenAIAnthropicGoogleDeepSeekxAIQwenMetaMistralZhipuMoonshotMiniMaxByteDanceKlingVidu
Pricing
Every frontier model, one balance
One API key, one balance and one dashboard for the whole model catalogue. Top up when you need to and spend it across any provider.
Free trial
Free credit on signup
$0starter credit
- Free credit the moment you sign up
- Run a real request in about 3 minutes
- OpenAI-compatible base URL
- Upgrade only when you ship
Enterprise
Volume, invoicing and support
Customtalk to us
- Volume discounts on every model
- Invoicing and annual contracts
- Dedicated routing and rate limits
- Priority support channel
Model prices
Published list prices, no markup games
Usage is drawn from your balance at the list price of each upstream model. Prices below are indicative list rates per 1M tokens - the console is the source of truth.
| Vendor | Model | Context | Input / 1M | Output / 1M |
|---|---|---|---|---|
| OpenAI | GPT-5 | 400K | $1.25 | $10.00 |
| OpenAI | GPT-5 mini | 400K | $0.25 | $2.00 |
| Anthropic | Claude Sonnet 4.5 | 200K | $3.00 | $15.00 |
| Anthropic | Claude Opus 4.1 | 200K | $15.00 | $75.00 |
| Anthropic | Claude Haiku 4.5 | 200K | $1.00 | $5.00 |
| Gemini 2.5 Pro | 1M | $1.25 | $10.00 | |
| Gemini 2.5 Flash | 1M | $0.30 | $2.50 | |
| DeepSeek | DeepSeek V3.2 | 128K | $0.28 | $0.42 |
| DeepSeek | DeepSeek R1 | 128K | $0.55 | $2.19 |
| xAI | Grok 4 | 256K | $3.00 | $15.00 |
| xAI | Grok 4 Fast | 2M | $0.20 | $0.50 |
| Qwen | Qwen3 Max | 256K | $1.20 | $6.00 |
| Zhipu | GLM-4.6 | 200K | $0.60 | $2.20 |
| Moonshot | Kimi K2 | 256K | $0.60 | $2.50 |
| MiniMax | MiniMax M2 | 200K | $0.30 | $1.20 |
| Mistral | Mistral Large 3 | 128K | $2.00 | $6.00 |
| Meta | Llama 4 Maverick | 1M | $0.27 | $0.85 |
| ByteDance | Seedance 2.0 | Video | per clip | per clip |
Quick start
From signup to a first call in 3 minutes
Register, copy a key, swap the base URL. Four steps stand between you and a working request.
Create an account
Sign up with a single click. Your starter credit lands in the console immediately.
Copy your API key
Open token management in the console, add a token and copy the key once.
Point your base URL
Swap the base URL in the SDK or dev tool you already use. Nothing else changes.
Ship the first call
Send one request, confirm the response and roll the key into staging.
Model catalogue
Direct routes to 400+ large language models
OpenAI, Anthropic, Google, DeepSeek, xAI, Qwen, Meta and more - one endpoint, one key, one bill.
OpenAI
24 models
Anthropic
14 models
21 models
DeepSeek
9 models
xAI
8 models
Qwen
26 models
Meta
18 models
Mistral
15 models
Claude CodeCodexCursoropencodeCherry StudioTraeOpenClawHermes
Workbench
Turn a prompt into something you can ship
The same key powers chat, image and video models, so a creative pipeline is one integration instead of five.
Video
Release night
ByteDanceSeedance 2.5
Video
Through the painting
ByteDanceSeedance 2.0
Video
Coffee shop detour
ShengshuVidu Q3 Pro
Video
Breakfast in the garden
KuaishouKling 3.0 Turbo
Video
Knights of the forest
xAIGrok Imagine Video
Setup guide
Pick your tool and copy the config
After you create a key, paste the base URL and model name into the tool you already use. Claude Code takes the host without /v1 - everything else takes the full path.
API key
Console → Token management → copy the keyBase URL
https://api.quickrouter.homesModel
claude-sonnet-4-5export ANTHROPIC_BASE_URL="https://api.quickrouter.homes"
export ANTHROPIC_AUTH_TOKEN="sk-qr-your-key"
claude "summarise this repository"Claude CodemacOS / Linux
npm install -g @anthropic-ai/claude-code
claudeClaude CodeWindows PowerShell
npm install -g @anthropic-ai/claude-code
$env:ANTHROPIC_BASE_URL="https://api.quickrouter.homes"CodexmacOS / Linux
npm install -g @openai/codex
codexCodexWindows PowerShell
npm install -g @openai/codex
$env:OPENAI_BASE_URL="https://api.quickrouter.homes/v1"Trusted by developers
Teams keep one gateway in front of every model
From solo builders to product teams shipping with OpenAI, Claude, Gemini, DeepSeek and Grok every day.
Alex ChenFull-stack engineer
Li MingAI product lead
Emily ZhangTechnical PM
Wang YihangAutomation developer
Lin ZhouAI architect
Chen RuoxiIndie developer
Daniel WangBackend engineer
Jessica MartinezHead of design
Zhao YuSaaS founder
Robert TaylorSoftware architect
FAQ
Questions developers ask before the first call
Everything about keys, base URLs, billing and compatibility with the OpenAI protocol.
Can the free credit run a real test?+
Yes. Every new account gets starter credit that is enough for a first round of API calls, so you can validate the base URL, the key and your SDK wiring before you top up.
Who do I contact when a tool will not connect?+
Mail support@quickrouter.homes with the tool name, the exact error and the request id when you have one. Connection problems are usually a trailing /v1 or a key that was copied with whitespace, and both take a minute to fix.
What should I put in the base URL field?+
Most tools take https://api.quickrouter.homes/v1. Claude Code is the exception: it expects https://api.quickrouter.homes without the /v1 suffix. Each card in the setup guide above shows the value to paste.
How do I configure Claude Code, Cursor or Codex?+
Open the setup guide on this page, pick your tool and copy the API key, base URL and model name into the fields listed on the card. The snippets are written for a clean install.
What is the first thing to do after signing up?+
Sign in, open token management and create an API key. Copy it once and store it in your secret manager - the console only shows the full key at creation time.
Is the gateway compatible with the OpenAI protocol?+
Yes. Requests and responses follow the OpenAI chat completions shape, so existing code usually needs only a new base URL and API key. Streaming, tools and vision inputs are supported on models that expose them.
How is usage billed?+
Usage is priced at the list price of each upstream model and drawn from your balance. Nothing is rounded up, and the console shows per-request detail so you can reconcile spend against your own logs.
What is an LLM API gateway?+
A gateway keeps one endpoint, one key and one billing relationship in front of many providers. Your code stops depending on a single vendor, and switching models becomes a one-line change instead of a new integration.
Start free and make the first call in 3 minutes
Create an account, copy your API key and point your base URL at QuickRouter.AI. No monthly fee, no card required to test.