Precision File Search

Input—per 1M tokens

Output—per 1M tokens

Context—tokens

WeightsClosed

About

Precision File Search is ranked #9 of 59 in desktop search software on Inferse. It runs on Linux, macOS, Windows. There is a free plan.

Compared on desktop search software

Free plan: Yespfs-ai.github.io
Supported platforms: Windows, macOS, Linuxpfs-ai.github.io
Index locations: localpfs-ai.github.io
Content search: Yespfs-ai.github.io

Facts

What it does: Precision File Search is a desktop application for searching, classifying, and understanding local files.pfs-ai.github.io · 3 Oct 2026
Search modes: It combines high-speed classic search, semantic search using a RAG pipeline, and natural-language questions with summarized answers.pfs-ai.github.io · 3 Oct 2026
Classification: Its trainable machine-learning model can classify scattered documents into user-tailored categories.pfs-ai.github.io · 3 Oct 2026
Local operation: The maker says the application and AI models can run entirely on the local machine, and that it works offline without cloud dependency.pfs-ai.github.io · 3 Oct 2026
Privacy: The maker says PFS never uploads files or search queries and that personal data is not sent to external servers.pfs-ai.github.io · 3 Oct 2026
Security: The maker lists path-traversal prevention, XSS sanitization, and prompt-injection hardening as security measures.pfs-ai.github.io · 3 Oct 2026
Languages: The site lists support for 18 languages, including English, Arabic, Chinese, French, German, Hindi, Japanese, and Spanish.pfs-ai.github.io · 3 Oct 2026
Intended users: The project README names researchers, students, developers, legal and business professionals, writers, and people organizing large local document collections as intended users.github.com · 3 Oct 2026
Windows download: The project README directs users to download a Windows installer with an .exe extension.github.com · 3 Oct 2026
AI model setup: The release page says an internet connection is needed on first run to download multilingual AI models, after which subsequent runs can be fully offline.github.com · 3 Oct 2026
AI integrations: The README says PFS supports OpenAI-compatible inference URLs, including online providers such as OpenAI and Together.ai and local engines such as Ollama and LM Studio.github.com · 3 Oct 2026
Hardware requirements: The README says the core platform runs on most modern consumer hardware with at least 8 GB of RAM; local LLM use is listed as needing a dedicated GPU, with at least 4 GB VRAM for smaller models.github.com · 3 Oct 2026
License: The project README states that PFS is licensed under the Mozilla Public License 2.0 (MPL-2.0).github.com · 3 Oct 2026
Business services: The README offers organizations custom feature development, enterprise integrations, priority support and maintenance, on-premise deployment consulting, and AI/RAG consulting.github.com · 3 Oct 2026
Maintainer: The README says PFS was developed by Aran Kazemi and describes him as an IBM Certified AI Engineer.github.com · 3 Oct 2026