# prabanta-dev/janas

> Janas is a family of AI tools in C with no dependencies. Janas-LLM runs large MoE models (up to 80B, 48 GB) on a 32 GB laptop by streaming experts from NVMe SSD, and outperforms llama.cpp on all 13 tested models.

- Magnitude: 6.0 out of 10 — Early signal
- Stars: 105 total · +51 stars measured 2026-10-13, 01:25–04:13 UTC, ≈ 143 by evening
- Star trust: star growth looks organic
- Category: Language models · Language: C · License: GPL-3.0 · Created: 2026-09-20 · Last push: 2026-10-12
- GitHub: https://github.com/prabanta-dev/janas · Page: https://gitnova.dev/en/r/prabanta-dev/janas

## Useful for

- Run Qwen3-Next-80B on a 32 GB laptop via janas-chat
- Measure your machine's inference speed with janas-bench
- Compare Janas-LLM and llama.cpp performance on local models

## Why it’s here

- Star-counter measurements on 2026-10-13 (UTC), 01:25–04:13: 54 → 105 stars (+51). This is the change over that interval.
- Estimated end-of-day forecast: about +143 stars, using observed gains and the previous day.
- Before this, the repository barely got any stars — about 0 per day.
- The repository is 23 days old and already has 105 stars.
- Hacker News: “Janas: Run LLMs larger than RAM by streaming experts from SSD” — 66 points, 6 h ago.

## Star trust

Star growth looks organic.

Star-trust labels are heuristics based on the repository’s behavior, not a check of every stargazer.

## Numbers

- Forks: 8
- Issues and pull requests: 14
- Watchers: 1
- Average over the last week: 31 per day
- Usual pace: 1 per day
- Stars in the last hour (measured): 16
- Latest release: v2026.10.12.1 (2026-10-12)

## Stars per day, last 24 days (oldest → newest, today is partial)

2026-09-20 … 2026-10-13: 5, 4, 6, 3, 3, 1, 1, 2, 1, 3, 0, 0, 0, 0, 2, 0, 1, 0, 0, 0, 0, 0, 75, 51

## Hacker News

- Janas: Run LLMs larger than RAM by streaming experts from SSD — 66 points, 25 comments: https://news.ycombinator.com/item?id=50063717

## Spotted in now

- Spotted on Hacker News

## More in this category

1. **open-webui/open-webui** — 8.0 · Breakout · Language models · Python · +228 stars measured 2026-10-13, 00:02–04:13 UTC, ≈ 988 by evening
   Open WebUI is a self-hosted web interface for AI models via Ollama and OpenAI-compatible APIs, with RAG, plugins, agents, and team collaboration.
   Full card: https://gitnova.dev/en/r/open-webui/open-webui.md
2. **lyogavin/airllm** — 7.6 · Breakout · Language models · Jupyter Notebook · +158 stars measured 2026-10-13, 00:01–04:13 UTC, ≈ 896 by evening
   A library for inference and training of large language models on low-VRAM GPUs: it streams weights layer by layer from disk, letting 70B–671B models run on 4–12 GB of VRAM without quantization or distillation.
   Full card: https://gitnova.dev/en/r/lyogavin/airllm.md
3. **microsoft/BitNet** — 6.7 · Breakout · Language models · C++ · +69 stars measured 2026-10-13, 00:02–04:13 UTC, ≈ 296 by evening
   Official inference framework for 1-bit LLMs (BitNet b1.58) with optimized kernels for CPU and GPU. Runs ternary models fast and losslessly, including a 100B model on a single CPU.
   Full card: https://gitnova.dev/en/r/microsoft/BitNet.md
4. **mw00/project-maya** — 5.5 · Early signal · Language models · C++ · +22 stars measured 2026-10-13, 00:02–04:13 UTC, ≈ 131 by evening
   Local engine, server and dashboard to run the 321B MoE model GLM-5.3-Flash on one or two NVIDIA GPUs, offloading experts to RAM and NVMe.
   Full card: https://gitnova.dev/en/r/mw00/project-maya.md
5. **Niko1221/Strata** — 4.8 · Steady · Language models · C++ · +181 stars measured 2026-10-13, 00:02–04:14 UTC, ≈ 950 by evening
   A C++ local inference engine that runs the 125B MoE model Qwen3.8-Flash-Next on a regular PC with one NVIDIA GPU (12-24 GB) and 64 GB RAM, exposing an OpenAI/Anthropic-compatible API on localhost.
   Full card: https://gitnova.dev/en/r/Niko1221/Strata.md

---

Magnitude (0–10) measures how fast and how unusually interest in a repository is growing right now. It is not a quality score.
Days are UTC. “So far today” is a fact; “expected by the end of the day” is a forecast.
Summaries and use cases are written by an LLM (DeepSeek V4.1 Flash) from the README and may be inaccurate: verify specific claims (benchmarks, speed, hardware) in the repository itself.
Data as of 2026-10-13 04:24 UTC, updated every 30 minutes.