Sonjari
Local AI agent for Windows

Your PC does the work. You get the proof.

Sonjari is a free local AI agent that works right on your own PC, on a model running on your GPU. It researches the web in your real Chrome, builds Excel and Word documents, and reads and fixes code, and it shows you what it actually read. It works on its own, takes on tasks handed off by Claude Code, Codex or Cursor, and can take work from your phone over Discord. No API keys, no subscription, no usage meter.

  • No meter on anything that runs on your PC
  • Picks up when Claude Code or Codex hits its limit
  • Every run leaves receipts you can check

Windows 10/11 · NVIDIA GPU · Free on the Microsoft Store

Sonjari's computer 0.0 s
교보문고에서 '파이썬' 책 3권을 찾아 제목·판매가·적립 포인트를 엑셀로 정리해 줘 “Find 3 books for ‘Python’ on Kyobo Book Centre and put title, price and reward points in a spreadsheet.”
0 s25 s50 s75.1 s
Plays at 4× · real times shown
What you asked for0/3

    Redrawn from the record of a real run · 28 Sep 2026, 23:51 KST · one RTX 3090 PC · 9 steps, 0 failed · run 026df5bd

    Watch a run

    One sentence in. A real file out.

    A real run, sped up about 9× · RTX 3090 · Qwen3.8 27B · 20 Oct 2026 · it took 2 min 46 s

    With Claude Code: it hands the research to Sonjari

    4× speed · Claude Code on the left, Sonjari on the right · the pages are read on this PC and only the answer and sources go back · in this run Claude also points out a Word file Sonjari saved that nobody asked for (fixed in 0.4.28)
    The app itself

    Same task, in the real app

    The answer on the left, the file it made, and the requirement check at 3/3. On the right, Sonjari’s computer: the step ruler and the page it read last.

    Sonjari app after the Kyobo task: an answer table of three Python books with price and points, a created spreadsheet file, and a panel showing 8 steps and 3 of 3 requirements met.
    Screenshot of the app · same request run again on 30 Sep 2026 on a clean test profile · 8 steps, 3/3 · this screenshot shows the Korean interface
    Three promises, checked like a task

    What we promise, and how it's going

    A promise without a record is a slogan. Each one below has the number we measured, how many runs, and when. When we break one, it stays on this page.

    • It won't take your mouse.

      Kept

      Chrome used to jump in front of your work when a task started. It did in 7 of 8 runs. After the fix, 0 of 8.

      n=8 · 2026-09-21
    • It says so when it couldn't do it.

      Broken once

      Honesty bench: asked across 3 routes with 3 questions, it pretended to have read something 0 times out of 9.

      n=9 · 2026-09-30

      28 Sep 2026: in one run it reported a task as done after copying a recommendation banner, with no warning. We fixed the checker the same night. The rematch that night: 0 false completions in 6 runs.

    • No meter on what runs on your PC.

      Kept

      The model runs on your GPU. Web search runs in your own Chrome, so there is no search API key and nothing to top up. We don't spend your Claude or ChatGPT subscription behind your back.

      How it's built, not a benchmark
    Why people add it

    It works next to the AI you already pay for

    When Claude hits its limit, keep going

    Tell Sonjari you're out. It finds where Claude Code or Codex stopped, puts that work on a new branch and carries on locally.

    Spotting the stop: caught 52 of 52 stopped sessions, 1 false alarm in 458 ordinary messages. How often the handed-over work gets finished: not measured yet.

    Hands for your frontier model

    Connect it to Claude Code, Codex, Cursor or the Claude app over MCP. The frontier model plans; Sonjari does the reading and clicking on your PC and sends back the result.

    On Korean shopping sites, Claude Code's own cost was lower with Sonjari in 9 of 9 paired runs ($1.71 vs $3.69 alone), and correct rows did not drop. It took about 4× longer. 27 rows per arm, 2026-10-15.

    Your files stay on your disk

    Excel, Word and HWP files are read and written on your PC. Nothing leaves it unless you send it, and a bug report shows you exactly what it will send first.

    The run above wrote its spreadsheet to your Documents folder and re-checked it: 4 rows.

    Before you install

    What it can't do yet

    • Windows only. Windows 10 or 11.
    • It needs an NVIDIA GPU. 6 GB runs a small 4B model that works best next to the AI you already use. 12 GB runs a 9B, and 24 GB runs the 27B we test on. AMD and Intel cards are not measured yet.
    • It is slower than the cloud. On the same price research, Claude Code alone took 96 s and with Sonjari 401 s (median). On one coding task Hermes Agent was about 2× faster.
    • The English interface is new. It shipped in October 2026. Tell us where the wording is off.
    • The numbers are still small. On English sites a 12 GB-class model finished GitHub trending in 7 of 7 runs and arXiv in 4 of 4. One of those runs put a wrong number in the table.
    • This is the first public release. Expect rough edges and tell us about them.
    Field log

    Every run we measured, including the ones we lost

    Newest first. Each line says what we tried, how many runs, and what happened.

      Free on the Microsoft Store

      Try it on your PC.

      It installs in one click from the Microsoft Store and updates itself. The source is on GitHub under Apache-2.0. We read every report.

      Questions, or something went wrong? Write to us, or open an issue on GitHub.

      Source code: github.com/sonjari/sonjari · Apache-2.0