#NPU

2025-05-03

There are #Framework13 models now with #AMD CPUs! This is getting interesting for many reasons! 👀

1. The AMD Ryzen AI CPUs come with a #NPU based on #XDNA from the #Xilinx acquisition. In practice #AMDXDNA only made it into Linux 6.14 (e.g. in the very recent Fedora 42) and there is no end user support as of today making the NPU dead hardware as of today and probably most of 2025.

2. The AMD Ryzen CPUs come with AVX512 and that's working end to end unlike whatever the f #Intel is doing there.

Framework 13 CPU options:

ADM Ryzen AI 5/7/9
Farooq | فاروقfarooqkz@cr8r.gg
2025-04-27

After my #wake_word_detection #research has delievered fruits, I have plans to continue works in the voice domain. I would love if I could train a #TTS model which has #British accent so I would use it to practice.

I was wondering if I could do the inference on #A311D #NPU. However, as I am skimming papers of different models, having inference on A311D with reasonable performance seems unlikely. Even training of these models on my entry level #IntelArc #GPU would be painful.

Maybe I could just finetune an already existing models. I am also thinking about using #GeneticProgramming for some components of these TTS models to see if there will be better inference performance.

There are #FastSpeech2 and #SpeedySpeech which look promising. I wonder how much natural their accents will be. But they would be good starting points.

BTW, if anyone needs opensource models, I would love to work as a freelancer and have an #opensource job. Even if someone can just provide access to computation resources, that would be good.

#forhire #opensourcejob #job #hiring

#AI #VoiceAI #opensourceai #ml #speechrecognition #speechsynthesis #texttospeech #machinelearning #artificialintelligence

MSvanamsvana
2025-04-19

Chip manufacturers like AMD and Intel started talking a lot about NPUs. They are power-efficient more power-efficient, but you may actually get better performance if you run AI models on an integrated GPU:

- pcworld.com/article/2199565/ai
- youtube.com/watch?v=a9NprGqBr54

Hacker Newsh4ckernews
2025-04-13

AMD NPU and Xilinx Versal AI Engines Signal Processing in Radio Astronomy (2024) [pdf]

git.astron.nl/RD/acap/-/raw/ma

#2024

ArmSoM_Officialarmsom_jackson
2025-04-11

ArmSoM boards (RK3576) CM5&Sige5 have been successfully tested with Ultralytics projects (e.g., YOLO11n) ,support ​​Ultralytics YOLO11n​​ with ​​Rockchip RKNN​​!
🔥 Achieve ​​low-latency, high-efficiency AI inference​​ for edge applications like industrial automation, smart retail, and robotics.
🔗 Dive into benchmarks & deployment:docs.ultralytics.com/integrati
Stay tuned!See how it goes!
@ultralytics @GlennJocher

Lup Yuen Lee 李立源lupyuen@qoto.org
2025-04-10
GNU/Trinukso ۞trinux@masto.ai
2025-04-10

Yo llevo veinticinco años usando exclusivamente software libre pero no soy un talibán. Eso, de entrada. Me gustaría preguntar / debatir acerca de #Recall.

Visto desde fuera parece una característica muy flipante. He leído criticas por lo de las capturas de pantallas y tal. Mi pregunta es: ¿acaso no se hace eso en local?

He leído que te requiere una #NPU. Aunque fuera localmente. Quién tiene cojones de fiarse de #Microsoft. Cómo va afectar a GNU/Linux las NPU. Para qué se pueden usar.

2025-04-10

[Перевод] Всё про инференс на Sophon NPU

В этой статье мы поговорим про ML на базе плат Sophon. Наверное это один из производителей которые набрали больше всего популярнсти в AI последнее время. Я расскажу как они соотносятся с другими платами на рынке (Jetson, RockChip, Hailo, TI, etc.). Расскажу как подготовить сети для работы на платформах, покажу ограничения (что пока нельзя сделать, ограничения по скорости, и.т.д.).

habr.com/ru/companies/recognit

#Sophon #NPU #Jetson #RockChip #ML #Edge_Inference #Computer_Vision #LLM #VLM #Qwen

2025-04-10

Native #NPU support is coming to #openSUSE!
Thanks to the Intel AI PC DevKit & #OpenVINO, #openSUSE will natively harness #Intel NPU power; no extra installs needed! Get ready for lightning-fast local #AI processing right from your #Linux system! 🦎💚 innovatorsforopensuse.org/2025

2025-04-10

Just learnt that AMD NPU support on Linux is WIP. Not time yet for the AMD RYZEN AI 300 / Linux / Ollama setup. #npu #linux

2025-04-01

Expanding Copilot+ PC experiences across AMD, Intel and Snapdragon powered devices. blogs.windows.com/windowsexper... #windows #windows11 #copilotpluspc #npu #ai

Expanding Copilot+ PC experien...

2025-04-01

📱💡 Descubre los smartphones con la NPU más potente de 2025. ¡Un salto en rendimiento y funciones gracias a la inteligencia artificial! 🤖 #Tech #Smartphones #NPU
androidayuda.com/android/gener

Thor A. Hoplandhopland@snabelen.no
2025-03-18

This apologia being said, the fact is that #AI shouldn't be here - yet.

In a responsible world this would be an idea on paper until we've reached a point where we could develop a #NPU to facilitate #LLM or #generativeAI - as an academic project that could reach maturity in like a 100 years

But since #capitalism is the #socioeconomic equivalent of shooting cocaine straight into the vein, what we see is irresponsible execs - who are destroying #acedemia - while slamming their foot on the pedal.

Thor A. Hoplandhopland@snabelen.no
2025-03-12

And pop goes the bubble. The plagiarism machine is gonna go vrrrooom, but as a general utility? We'll see how much use people will have for their #NPU chips. Will it become one of those accelerators that will be abandoned and lost to time...?

"Accelerators"? Again; DEAR GOD I'm old.

New survey suggests the vast majority of iPhone and Samsung Galaxy users find AI useless – and I’m not surprised | TechRadar
techradar.com/phones/new-surve

2025-03-11

Running DeepSeek-R1 14B on a CoPilot+ Snapdragon PC in Airplane Mode (using the NPU!) | with Scott Hanselman. www.youtube.com/watch?v=jEXC... #ai #deepseek #npu #copilotpluspc #arm64

Running DeepSeek-R1 14B on a C...

Client Info

Server: https://mastodon.social
Version: 2025.04
Repository: https://github.com/cyevgeniy/lmst