Meruva Gateway · for developers
One key to Qwen, DeepSeek and Kimi.
One OpenAI-compatible endpoint for Qwen, DeepSeek, Kimi and more. Built for experiments, side projects and code you're happy to share.
Gateway requests are processed by our upstream provider in mainland China. Details on Trust.
What you get.
Open models, one bill, and the freedom to leave whenever you like.
01
One endpoint
OpenAI-compatible, plus Anthropic-style where supported. Switch models by changing one string.
02
Works with your agents
Use Claude Code-style agentic tools. We'll help you set up cc-switch so your coding agent runs on the model you choose.
03
Clear costs
Usage per key and per model, so you know what every call cost. Pay by card through Stripe. No contracts.
How it works
Three steps to your first call.
- Pay. Sign up and top up by card through Stripe.
- Get your key. Create a key in the dashboard.
- Connect your tools. Point your app or coding agent at Meruva, then watch usage and spend in the dashboard.
Open models only
Qwen, DeepSeek, Kimi and more.
We only offer open-weight models you could run yourself. No US models (why?). Setup guide coming soon.
Where a request travels
01Your appAny OpenAI-compatible client
02MeruvaKeys, usage and billing
03Upstream providerModel inference · see Trust