Xuan-Son Nguyen's banner
Xuan-Son Nguyen's profile picture

Xuan-Son Nguyen

@ngxson • 6,843 subscribers

Engineer @huggingface

Shorts

Fruit fly brain as LLM 🤷‍♂️🤷‍♂️

Fruit fly brain as LLM 🤷‍♂️🤷‍♂️

34,284 просмотров

Qwen3.6-27B running 100% on WebGPU. Not the best speed but still 😁

Qwen3.6-27B running 100% on WebGPU. Not the best speed but still 😁

53,142 просмотров

Videos

LIVE
1.2k

Anya Rossi

sweetdream.ai

SweetDream.aiSponsored•Livecam
Streaming Now

Watch Anya Live

Anya is streaming live right now! Join her private show and enjoy exclusive content.

HD live stream
Exclusive private shows
1.2k viewers online

Free preview available • Premium content

Real-time webcam demo with Hugging Face SmolVLM and ggml llama.cpp server. All running locally on a Macbook M3
0:24
ngxson's profile picture

Real-time webcam demo with Hugging Face SmolVLM and ggml llama.cpp server. All running locally on a Macbook M3

Xuan-Son Nguyen

974,221 просмотров • 1 год назад

You heard about LLM inference on WebGPU, but what about... finetuning LLM on WebGPU? 🤯 I put together a super earlier PoC that proves it's possible, backed by llama.cpp / wllama 😁 Working on LoRA next...
0:29
ngxson's profile picture

You heard about LLM inference on WebGPU, but what about... finetuning LLM on WebGPU? 🤯 I put together a super earlier PoC that proves it's possible, backed by llama.cpp / wllama 😁 Working on LoRA next...

Xuan-Son Nguyen

19,426 просмотров • 8 дней назад

llama.cpp is likely the first LLM runtime in the world to allow "interrupt" reasoning without stopping the whole response. We also added a small "skip" button on the Web UI, the model gives the final response as soon as you click the button. The response is no longer bound to reasoning budget!
0:31
ngxson's profile picture

llama.cpp is likely the first LLM runtime in the world to allow "interrupt" reasoning without stopping the whole response. We also added a small "skip" button on the Web UI, the model gives the final response as soon as you click the button. The response is no longer bound to reasoning budget!

Xuan-Son Nguyen

31,511 просмотров • 3 месяцев назад

Introducing: the new llama-cli 🦙🦙 > Clean looking interface > Multimodal support > Conversation control via commands > Speculative decoding support > Jinja fully supported
0:26
ngxson's profile picture

Introducing: the new llama-cli 🦙🦙 > Clean looking interface > Multimodal support > Conversation control via commands > Speculative decoding support > Jinja fully supported

Xuan-Son Nguyen

21,343 просмотров • 9 месяцев назад

Больше нет контента для загрузки