Back to browse
Chuk.chat – Privacy-first AI chat with multiple models

Chuk.chat – Privacy-first AI chat with multiple models

by chukfinley2·Feb 14, 2026·2 points·0 comments

AI Analysis

●●SolidDark HorseSolve My Problem

Open models + flat pricing, but ChatGPT/Claude already own privacy-conscious users.

Strengths
  • Transparent open-weight-only approach eliminates black-box concerns others face
  • Flat €20/month sidesteps token-anxiety UX pain of pay-as-you-go models
  • Cross-platform native apps (Windows, macOS, Linux, Android) match Claude Desktop distribution
Weaknesses
  • Privacy-first positioning crowded (Perplexity, HuggingChat, local alternatives exist)
  • No evidence of differentiated model performance vs. competitors' open-weight offerings
Category
Target Audience

Users concerned about data privacy, researchers, developers wanting transparent AI

Similar To

Perplexity AI · HuggingChat · LocalAI

Post Description

I built Chuk Chat because I wanted AI chat that doesn't treat my data as a product.

It runs only open-weight models: DeepSeek V3.2, Qwen3, Kimi K2.5, MiniMax M2.5, GLM 5. Encrypted chats, no tracking, fully deletable data.

€20/month flat, no token counting.

Available on Web, Windows, macOS, Linux, Android. iOS coming soon.

https://chuk.chat

Similar Projects

AI/ML●●Solid

I'm running ML models for PDF layout analysis entirely in the browser

Runs full layout detection — PP-DocLayoutV3 for blocks/titles/tables and Microsoft's table-transformer for cell structure — entirely in the browser via ONNX Runtime (WASM/WebGPU), so documents never leave the tab. Impressive technical work, but the models are ~60MB and inference is CPU/GPU-heavy, so expect crashes on phones and slow analysis on older machines. The plugin-first approach (React/Svelte/Vue) and open-source repo make it a very usable proof-of-concept for privacy-first PDF tooling.

WizardryBold Bet
bobsingor
304mo ago