Breaking

0xSojalSec's LLMs-local List Catalogs Platforms, Engines, and UIs for Running LLMs on Local Hardware

September 2, 2026 at 11:37 EDT

Open Source
Foundation Models

A GitHub repository called LLMs-local, maintained by 0xSojalSec, lays out a structured directory of platforms, inference engines, and user interfaces for running large language models on local hardware instead of cloud APIs. As of September 2, 2026, the public catalog has 1,095 stars and 126 forks, and its README was last pushed on June 10, 2026.

September 2, 2026 · GitHub

Mapping the local LLM stack

A public catalog groups five platforms, 15 inference engines, and five chat interfaces for running open-weight models on your own hardware.

1,095

GitHub stars

Inference engines

5 + 5

Platforms and UIs

126

Forks

How the catalog is weighted

Each block is one listed tool. Engines outnumber platforms and chat UIs three to one.

Platforms

Engines

User interfaces

Typical local stack

Platforms discover models, engines run the weights, and UIs sit on an API.

Platform

Discover and download

→

Engine

Load GGUF and generate

→

User interface

Chat, IDE, or web UI

Memory floors, drawn to scale

Each block is 4GB. Hardware, not the catalog, decides which model class loads.

8GB

3B-class RAM

16GB

7B-8B RAM

24GB

30B-class VRAM

What helps

Layered sections and live star badges give a fast way to pick a starter stack: Ollama, LM Studio, or Jan.

What lags

The README last changed on June 10, 2026. Eight pull requests from tool authors remain unmerged, and older roundups go stale.

Continue reading

The rest of this article is for AI News Blitz readers. Choose an option below to keep reading.