Runs 100% on your machine · no cloud

Your machine.
Your model.
Your AI workshop.

Saient is a local AI desktop app — chat, a real coding agent, image generation, text-to-speech, and LoRA training, all running on your own GPU. No API keys, no subscriptions, nothing leaves your computer.

Tauri + Svelte tinyq4 GGUF engine GPU-accelerated
[>_<] /|#|\
  / \
saient — agent · ~/projects/snake
What's inside

One app. Five studios. Zero cloud.

Everything a local AI workstation needs, in a single native window — each tool talks to the same on-device model.

>_

Chat

Talk to any local GGUF model. Renders live HTML artifacts in a side pane — ask it to build a tool or a game and watch it run.

GGUFlive artifacts

Saient — the coding agent

A real agent with a PTY terminal and a tool-use loop (read · ls · write · edit · bash). It plans, writes files, runs commands, and fixes its own errors — on your local model, in a sandboxed workspace.

Ctrl · Shift · K — toggle Saient

Image Gen

SDXL text-to-image with LoRA support, schedulers, live progress, and an optional face-detail pass.

SDXLLoRA

Text-to-Speech

Natural, expressive speech via Kokoro voices — generated entirely offline.

Kokoro

LoRA & Merge

Train your own SDXL LoRAs and merge checkpoints, with a built-in dataset cleaner.

trainmerge

One engine

Powered by tinyq4, a from-scratch GGUF inference engine — no llama.cpp, no Python server you have to babysit.

tinyq4native
The agent

It doesn't just answer.
It does the work.

Open the Agent screen, type saient, and hand it a goal. It runs an autonomous plan→act→verify loop until the job is done.

  • planBreaks the goal down into concrete steps and a JSON plan you can edit.
  • actReads, writes, edits and runs inside a sandboxed workspace.
  • verifyChecks its own output, catches errors, and re-plans until it succeeds.
  • safeWrite mode is off by default; destructive tools ask before running. Every action is logged to an audit trail.
planner · autonomous run
goal › add a /health endpoint + a test
───────────────────────────────
planning… 3 steps
read server.py
edit server.py +12 −0
write test_health.py
bash pytest -q
2 passed — goal achieved
Local-first, by design

Your data never leaves the building.

0
bytes sent to the cloud
0
API keys required
100%
offline after setup
generations, no metering
[>_<]
 /|#|\
  / \

Saient is ready.

Download the desktop app, point it at a GGUF model, and you've got a full AI workshop that answers to nobody but you.