01 / SELF-HOSTED
Your server.
Keep the heavy lifting on your computer. Connect to LM Studio, Ollama, vLLM, or another OpenAI-compatible endpoint.
Connect a server YOUR AI, A LITTLE CLOSER
Your AI.
In your pocket.
Your server, your on-device models, your cloud APIs.
One thoughtful chat client for iPhone and iPad.
Buy once. No Vanta account. No Vanta subscription.
BUILT AROUND YOUR CHOICE
Self-hostedOn-deviceCloud APIs
VANTA / IPHONEYOUR MODELS, WITH YOU
A familiar model.
A better place to chat.Ollama · Self-hosted
A better place to chat.Ollama · Self-hosted
Vanta is an AI chat client for iPhone and iPad, built by Ezel Bayraktar. Connect to your own server, run supported models on-device, or bring your cloud API keys.
Compare local AI setupsA PLACE FOR ALL YOUR MODELSThree ways in.
Three ways in.
One place to think.
Use what you already have.
Try something new when you need it.
02 / ON-DEVICE
Your device.
Download a supported GGUF or MLX model and take it offline. Or use Apple Intelligence when your device supports it.
Go on-device03 / CLOUD
Your provider.
Bring an OpenRouter key or add your own compatible cloud endpoint. Choose the model that suits the moment.
Bring your API keyMORE ROOM FOR THE GOOD STUFFBring the question.
Explore the features Bring the question.
And everything behind it.
Make your files
part of the conversation.
Ask about your PDFs. Add photos with a compatible vision model. Keep the context close to the question.
Work with your files
A second opinion.
And a third.
Put multiple models in one conversation. Give them roles, compare their answers, and find a different angle.
Meet your model panel
Tools that open doors.
Connect MCP servers or use research tools with compatible models.
Connect your toolsKeep your own context.
Save prompts, configure memory, and manage long conversations.
Make it yoursFeels at home on iOS.
Readable Markdown, code highlighting, voice input, and themes.
Look insideYOUR CHOICE
ALL THE WAY THROUGH.
ALL THE WAY THROUGH.
KNOW WHERE YOUR WORDS GOPrivacy starts
Privacy starts
with a clear answer.
Offline models run on your device. Self-hosted chats go to your server. Cloud chats go to your chosen provider. External tools have their own data flows.
Choose the setup that fits your conversation.
See exactly how data movesFROM “WHAT IF” TO YOUR FIRST CHAT
All documentation Start with your setup.
What is Vanta?
Vanta is an AI chat client for iPhone and iPad. Connect to your own server, run supported models on-device, or use cloud providers with your own API keys.
Does Vanta work with Ollama and LM Studio?
Yes. Vanta connects to their OpenAI-compatible server endpoints from iPhone or iPad. The model runs on your server; your device needs a network connection to it.
Read the Ollama setup guideCan I use Vanta without an internet connection?
Supported on-device GGUF and MLX models work offline after downloading. Online tools, cloud APIs, model downloads, and remote servers need a connection. Apple Intelligence also depends on system availability.
See on-device requirementsIs there a Vanta subscription?
No. Vanta is a one-time App Store purchase, with no Vanta account or subscription. Your cloud provider and other external services may charge separately.
See purchase detailsWhere do my conversations go?
On-device inference runs on your iPhone or iPad. Server and cloud requests go to the endpoint you choose. Enabled external tools and online services receive the data needed for your request.
Follow the data flow
MAKE SOME ROOM FOR YOUR AISame models.
New possibilities.
Download on theApp StoreFor iPhone & iPad · iOS 18 or iPadOS 18 or later