Zyekh AI OPEN SOURCE
GitHub
SELECT AI MODEL GATEWAY 8317
Enter Send • Shift + Enter New line • Model: Gemini 3.7 Flash
PROFILE & CUSTOMIZATION

User Profile Settings

U
Max 2MB (stored locally in browser)
These instructions are automatically attached to every conversation.
EMBED INTEGRATION

Embed AI Widget on Your Website

<script src="https://chat.zyekh.com/chat-widget.js" data-persona="tech_mentor" defer></script>
Attribute Default Description
data-persona cs_store tech_mentor, cs_store, companion, cloud_assistant
data-title Auto Widget header title
data-api https://api.zyekh.com AI inference gateway backend endpoint
SHARE CONVERSATION

Public Conversation Snapshot

This public link serves a static, search-engine indexed conversation snapshot with Schema.org markup viewable by anyone.

Generating link...
Post to X Telegram Threads
VISUAL CARD GENERATOR

Export AI Response as Image

High-resolution 1200x630 visual card generated client-side via HTML5 Canvas. Ready for X, Threads, Reddit, and LinkedIn.

ON-DEVICE WEBGPU AI

Initialize On-Device AI Model

You are activating a Local Neural Network LLM running 100% on your browser GPU without server connectivity.

SPECS & PRIVACY
• Model Weights: ~185 MB (Downloaded once & cached in IndexedDB).
• System Requirements: WebGPU-enabled browser (Chrome, Edge, Brave).
• Privacy: 100% Offline, Zero Telemetry, Zero Server Cost.
Downloading model weights... 0%