Find tools › Open-source projects › Keep data private

janas
Open-source project · by prabanta-dev
Run massive AI models on a normal laptop by streaming model weights directly from an NVMe drive.
Some setup
What it costs
Free and open source; you may pay for the AI model it uses.
Use it to
- Run large mixture-of-experts models on laptops with low RAM
- Stream model weights directly from fast NVMe SSD storage
- Benchmark local LLM prompt and token generation speeds
How to start
Download the release archive using
curl -LO https://github.com/prabanta-dev/janas/releases/latest/download/janas-linux-x86_64.tar.gzUnpack the archive and enter directory with
tar xf janas-linux-x86_64.tar.gz && cd janasLaunch the interactive chat interface by running
./janas-chat
Before you install
- Found automaticallyNot reviewed by a person yet. Check it before installing.
- License: GPL-3.0You are allowed to use it under these terms.
- Last updated 2026-11-18
More detail
Janas is an experimental, pure C inference engine designed to stream large mixture-of-experts (MoE) LLMs directly from an NVMe SSD so models larger than system RAM can run locally.
Reach for it when you want to run massive MoE models (such as an 80B model) on consumer hardware with limited RAM, reading weights from disk rather than keeping everything in memory.
Similar tools
Descriptions are written in plain English from each project's own documentation and refreshed by our agent; this one was last checked 2026-10-13. Spotted a mistake? · Open in Find tools · © 2026 Which AI Map