Private AI · by invitation

Think deeper.
Sharper than the mainstream.

Stryix is a private layer above consumer chat clients — direct access to frontier reasoning without the caps, queues, or product guardrails. Self-hosted infrastructure. Invite-only, for now.

Model per conversation
Each thread pinned to the intelligence that fits — reasoning, code, everyday work. Switch on demand, not by plan.
Redundant by design
Multiple providers behind every conversation. When one backend slows or falls over, the thread doesn't break — you don't even notice.
Owned end to end
Your conversations stay on your server. No training on your data, no third-party analytics, no shared infrastructure.
Durable by default
Full history, searchable and resumable. Custom system prompts per thread. Your thinking, preserved on your own infrastructure.
SSE streaming 1M token context JWT auth self-hosted invite-only direct API
plain 0 lines
plain
html live render · sandboxed
System Prompt — this conversation
Settings
Appearance
Theme
Personalization
Default system prompt
0
Applied to every new conversation. Per-conversation prompts override this.
Usage this month
total spend$0.00
queries0
input tokens0
output tokens0
Conversation
Export as Markdown
Export model
Project Brain
📁 Files 0
Storing…
No files yet. Add .py, .md, .html, .yaml…
📄 Documents 0
Indexing…
No documents yet. Add .pdf, .docx, .csv…
Session
Intelligence Panel
Live
Active model
V3
DeepSeek V3
Ready
deepseek api
—
tok/sec
This response
output — tokens
input context — tokens
Session cost
$0.0000
session starting…
this response —
today total —
Monthly budget
— —
— —
Session spend by model
No data yet — start chatting.
0
queries / mo
$0.00
month spend
0
input tokens
0
output tokens
premium
updates live