Bring your own key. Pay the provider, not a middleman.
With your own API key you choose the model, you see the real price, and nothing sits between you and the provider. VibeCode.surf is built that way from the start.
What that gets you
Any provider
Claude, OpenAI, OpenRouter, Gemini, Groq, DeepSeek, Z.ai (GLM), Moonshot Kimi, Qwen, Together AI, Mistral, xAI and any OpenAI-compatible server.
Prices in the picker
Each model shows its input and output price where the provider publishes it.
Cost per reply
Tokens, cached tokens and dollars under every reply, with totals per project and per model.
Thinking level
Choose how hard each agent thinks, from Off to Max, to trade speed and cost against depth.
Where your keys live
On your computer
Keys are stored locally and requests go straight from the app to the provider.
Nothing of ours in the path
Your VibeCode.surf account covers updates and your plan. Model requests never pass through our servers.
Or no key at all
Local models through Ollama and LM Studio need no key and cost nothing.
It handles trouble
A dropped connection is retried, and low credit shortens the reply instead of failing it.
What VibeCode.surf does not have yet
It is a beta. If you depend on any of these every day, keep your current editor alongside it for now.
- Autocomplete and go-to-definition from language servers
- Inline AI edits and tab completion inside the editor
- A debugger with breakpoints
- An extension marketplace
Good to know
Is one key enough?
Yes. A single router key such as OpenRouter gives access to many models. You can add more providers at any time.
Can different agents use different providers?
Yes. Each agent has its own model and provider.
Install it from your terminal.
One command installs the desktop app on macOS, Windows or Linux. Free during the beta; bring your own API key or a local model.
curl -fsSL https://vibecode.surf/install.sh | shirm https://vibecode.surf/install.ps1 | iexcurl -fsSL https://vibecode.surf/install.sh | sh