Unlimited private conversations.
One flat price.
Talk or whisper to DeepSeek V4 and Qwen 3.8 — frontier open-source AI, served from Switzerland. Switch between them mid-conversation. Sleep well.
Private by design
Your conversations are never stored nor logged on our servers — they live on your device, and only there. Only you can look back at them.
Hosted in Switzerland
Inference runs on Swiss servers via AI Router, with no prompt logging. Your words are processed, answered, and forgotten.
Open-source models
DeepSeek V4 and Qwen 3.8 — open weights anyone can inspect. No black boxes, and your chats train no one's next model.
Where everything lives
We don't own GPUs. The models are hosted in Switzerland by AI Router. They state that your prompts are processed in memory and never stored on their servers. They keep logs — IP addresses and timestamps — for 14 days against abuse. Your requests go to them from our server, not your browser: so the IP they log is ours, not yours. Your messages arrive under a key shared by all other cheaptalk.ai subscribers, with no name or email attached.
We also don't have a user database — your subscription lives at Stripe, and logging in just checks it's active. We don't store nor log conversations; they stay on your device, and only there. No trackers, no analytics, no ad pixels.
There is simply nothing here to leak, sell, or subpoena.
- Both models, unlimited conversations
- Streaming answers, model switch mid-chat
- Images, PDF & file uploads
- Voice input — speak instead of typing
- Read out loud — hear replies in a natural voice
- Chat in your browser — nothing to install
- Passphrase-encrypted export & import of your chats
- Cancel anytime