Lightnews — Scholar-powered news

Xuan Son Nguyen @ngxson.hf.co · 5d

Very nice touch, Gmail 😅

2

Xuan Son Nguyen @ngxson.hf.co · Aug 29

Link to article: blog.ngxson.com/building-my-...

Building My Smart Home - Part 2: ESPHome & RF433

Implementing an RF433 gateway using ESPHome and a custom Home Assistant add-on to manage multiple RF433 receivers for my smart home.

blog.ngxson.com

Xuan Son Nguyen @ngxson.hf.co · Aug 29

Part 2 of my journey building a smart home! 🚀

In this part:
> ESPHome & custom component
> RF433 receiver & transmitter
> Hassio custom addon

1

Xuan Son Nguyen @ngxson.hf.co · Aug 27

Link to article: blog.ngxson.com/building-my-...

Building My Smart Home - Part 1: Home Assistant

Designing a smart home system from electrical wiring to Home Assistant automations, using affordable devices and network-inspired architecture.

blog.ngxson.com

Xuan Son Nguyen @ngxson.hf.co · Aug 27

Just published a new article on my blog 🏃‍♂️

Building My Smart Home - Part 1: Plan, Idea & Home Assistant

Check it out!

1

Xuan Son Nguyen @ngxson.hf.co · Aug 14

Link here: huggingface.co/collections/...

Gemma 3-270m - a ggml-org Collection

Collection of models for Gemma 3-270m

huggingface.co

3

Xuan Son Nguyen @ngxson.hf.co · Aug 14

Kudos to Google and the llama.cpp team! 🤝

GGUF support for Gemma 270M right from day-0

1 5

Xuan Son Nguyen @ngxson.hf.co · Jul 21

Watch it here: www.youtube.com/watch?v=Qtzz...

Xuan Son Nguyen @ngxson.hf.co · Jul 21

Richy Mini and SmolLM3 are featured in Github's weekly news! 🚀 🚀

1

Xuan Son Nguyen @ngxson.hf.co · Jun 26

Gemma 3n has arrived in llama.cpp 👨‍🍳 🍰

Comes in 2 flavors: E2B and E4B (E means "effective/active parameters")

1

Xuan Son Nguyen @ngxson.hf.co · Jun 11

See you this Sunday at AI Plumbers conference: 2nd edition!

📍 Where: GLS Event Campus Berlin, Kastanienallee 82 | 10435 Berlin
👉 Register here: lu.ma/vqx423ct

Xuan Son Nguyen @ngxson.hf.co · Jun 3

✨✨ AIFoundry is bringing you the AI Plumbers Conference: 2nd edition — an open source meetup for low-level AI builders to dive deep into "the plumbing" of modern AI

📍 Where: GLS Event Campus Berlin, Kastanienallee 82 | 10435 Berlin
📅 When: June 15, 2025
👉 Register now: lu.ma/vqx423ct

1 2

Xuan Son Nguyen @ngxson.hf.co · May 15

Hugging Face Inference Endpoints now officially support deploying **vision** models via llama.cpp 👀 👀

Try it now: endpoints.huggingface.co/catalog

Xuan Son Nguyen @ngxson.hf.co · May 12

Check it out: github.com/ngxson/smolv...

GitHub - ngxson/smolvlm-realtime-webcam

Contribute to ngxson/smolvlm-realtime-webcam development by creating an account on GitHub.

github.com

Xuan Son Nguyen @ngxson.hf.co · May 12

Real-time webcam demo with @huggingface.bsky.social SmolVLM and llama.cpp server.

All running locally on a Macbook M3

1 2

Xuan Son Nguyen @ngxson.hf.co · Apr 25

Although we have A100, H200, M3 Ultra, etc

Still can't match the power of that Casio FX 😆

2

Xuan Son Nguyen @ngxson.hf.co · Apr 21

llama.cpp vision support just got much better! 🚀

Traditionally, models with complicated chat template like MiniCPM-V or Gemma 3 requires a dedicated binary to run.

Now, you can use all supported models via a "llama-mtmd-cli" 🔥

(Only Qwen2VL is not yet supported)

5

Xuan Son Nguyen @ngxson.hf.co · Apr 20

Learn more: blog.ngxson.com/introducing-...

blog.ngxson.com

1

Xuan Son Nguyen @ngxson.hf.co · Apr 20

Finally have time to write a blog post about ggml-easy! 😂

ggml-easy is a header-only wrapper for GGML, simplifies development with a cleaner API, easy debugging utilities, and native safetensors loading ✨ Great for rapid prototyping!

1

Xuan Son Nguyen @ngxson.hf.co · Apr 20

Someone at Google definitely had a lot of fun making this 😆

And if you don't know, it's available in "Starter apps" section on AI Studio. The app is called "Gemini 95"

1

Xuan Son Nguyen @ngxson.hf.co · Apr 20

Telling LLM memory requirement WITHOUT a calculator?

Just use your good old human brain 🧠 😎

Check out my 3‑step estimation 🚀

1 3

Xuan Son Nguyen @ngxson.hf.co · Apr 19

Google having a quite good sense of humor 😂

Joke aside, 1B model quantized to Q4 without performance degrading is sweet 🤏

1 2

Xuan Son Nguyen @ngxson.hf.co · Mar 31

Where to try? ggml-easy --> github.com/ngxson/ggml-...

GitHub - ngxson/ggml-easy: Thin wrapper around GGML to make life easier

Thin wrapper around GGML to make life easier. Contribute to ngxson/ggml-easy development by creating an account on GitHub.

github.com

Xuan Son Nguyen @ngxson.hf.co · Mar 31

Cooking a fun thing today, I can now load safetensors file directly to GGML without having to convert it to GGUF!

Why? Because this allow me to do experiments faster, especially with models outside of llama.cpp 😆

1

Xuan Son Nguyen @ngxson.hf.co · Mar 30

No vibe coding. Just code it ✅

Visit my website --> ngxson.com

2