It's #ROCm getting better? Yes

Will you still use #CUDA? Yes.

https://www.youtube.com/watch?v=wCBLMXgk3No&t=933

What #AMD should focus on is to bring all of their SKU to use ROCm stable on all platforms. Currently that isn't possible, which is frustrating given their cards have more memory than #RTX at the same price.

GamersNexus is cooking up an absolutely amazing #GPU smuggling story, and making a movie out of it!

https://youtu.be/ltgyS8oJC8g?si=RpSe-1T_xFcqw0Zy

#FluidX3D is now 3 years out in the wild! This fun hobby project has turned into the largest #CFD software on #Github, with 4.6k Stars. Community feedback has been overwhelming.

Since last year I've enabled using full multi-TB memory capacity on CPUs/iGPUs, added #GPU-accelerated force/torque summation, optimized vtk export, and lots of bug fixes, keeping the # of known bugs at 0.

More updates are planned, as well as a roadmap to commercial use. Let's keep growing!

https://www.youtube.com/watch?v=6enTZjO9fbQ

Gerade hat #AMD eine neue #Radeon-#Grafikkarte veröffentlicht. Die #RX7400 basiert als #RDNA3-#GPU auf dem Navi-33-Chip und ist mit einer TBP von 55 Watt sehr sparsam. Sie richtet sich wohl vor allem an OEMs. https://winfuture.de/news,152827.html?utm_source=Mastodon&utm_medium=ManualStatus&utm_campaign=SocialMedia

- Do you have a 48 GB RTX 4090 here?
- We can make one now.

Just an average store around the corner.

https://youtu.be/ltgyS8oJC8g?si=ocdjMAiA5jR-z-04

I wonder if the real win of the #AMD #Ryzen AI Max+ 395 isn't just being a really great desktop, but whether it's aimed at breaking the artificial stranglehold #Nvidia has on #GPU memory sizing.

Tell me a good reason to buy an $1,500 Ryzen AI MAX+ 395.

I'll put these sites as my absolute answer:

https://gaminglaptop.deals/
https://www.bestlaptop.deals/
https://www.reddit.com/r/LaptopDeals/

Ça vous dirait de découvrir nos travaux de recherche les plus avancés sur les impacts environnementaux de l'IA ?

Cette journée de partage, c'est le Hubblo Day. Cette 1ère édition est dédiée à la question de l’IA générative : des composants de calcul jusqu’à la transformation dans les organisations.

La billetterie est ici (100 places disponibles, prix libre) : https://my.weezevent.com/hubblo-day

Infos pratiques :...

The #AlmaLinux project has announced the availability of packages to enable native #NVIDIA driver support, including #CUDA and Secure Boot, for #AlmaLinux9 and #AlmaLinux10. https://almalinux.org/blog/2025-08-06-announcing-native-nvidia-suport/
#Linux #opensource #gpu

如何在 NVIDIA GPU 上以每秒 500+ 個 token 的速度運行 GPT OSS 120B
➤ 在上市首日，Baseten 如何透過實驗、除錯和基準測試，實現 GPT OSS 120B 的卓越效能。
✤ https://www.baseten.co/blog/sota-performance-for-gpt-oss-120b-on-nvidia-gpus/
本文詳細闡述了 Baseten 團隊如何在 NVIDIA GPU 上，針對 GPT OSS 120B 模型達成每秒超過 500 個 token 的尖端延遲和吞吐量。作者分享了從首次推理、修復相容性錯誤到優化模型配置的完整過程，強調了使用 TensorRT-LLM、TPU 協同運作、KV 快取路由和預測解碼等技術，並展示瞭如何在上市首日就為客戶提供最佳的性能體驗。
+ 看到 Baseten 團隊能在短時間內達成如此驚人的效能，真是令人印象深刻。特別是他們在 TensorRT-LLM 和 GPU 協同運作方面的深入研究，為業界樹立了標
#AI 模型效能 #GPU 優化 #大型語言模型 #OpenAI #Baseten

"NOPE!"

NVIDIA say no to adding backdoors and killswitches in their GPUs

https://www.gamingonlinux.com/2025/08/nvidia-say-no-to-adding-backdoors-and-killswitches-in-their-gpus/

The #Samsung Galaxy Z Fold 7 is powered by the Qualcomm Snapdragon 8 Elite for Galaxy #chipset. This is a special, overclocked version of Qualcomm's latest flagship processor, custom-tuned to deliver enhanced performance specifically for Samsung's devices. It features a faster #CPU, #GPU, and #NPU, providing superior speed for multitasking, gaming, and on-device #Samsung features like those found in Galaxy AI.
Source: https://faxty.com/samsung-galaxy-z-fold7-cpu

The #Samsung Galaxy Z Fold7 is equipped with the Adreno 830 #GPU, which is a part of the custom Snapdragon 8 Elite for Galaxy #chipset. This processor is a special version of Qualcomm's flagship chip, and the Adreno 830 GPU in the Fold 7 is specifically tuned to provide a significant performance boost for graphics and #AI, enabling smooth #gaming and advanced on-device AI features.
Source: https://faxty.com/samsung-galaxy-z-fold7-gpu

China’s First High-End Gaming GPU, the Lisuan G100, Reportedly Outperforms NVIDIA’s GeForce RTX 4060 & Slightly Behind the RTX 5060 in New OpenCL Benchmarks

Out with the… 7 months old, in with the new. The Intel GPU was supposed to be more stable, and lower power, but it turned out to be neither of those things, plus there was unsolvable tearing in X11. So back to AMD we go.

#NVIDIA #Engineer Now Co-Maintainer Of "NOVA" #OpenSource Rust #GPU Driver
NOVA opensource NVIDIA GPU effort the past year and 1/2 has been led by #RedHat engineer #DaniloKrummrich as the main NOVA-Core driver maintainer for #kernel code that continues being built out piece-by-piece for mainline tree.
Now joining him as a co-maintainer is NVIDIA engineer, #AlexandreCourbot. Name may ring a bell as he started out in 2011 working on Tegra graphics driver support for #Linux.
https://www.phoronix.com/news/NOVA-Core-Co-Maintainer

Linux Creator Linus Torvalds' Setup Reveals Surprising Budget AMD RX 580 GPU

#BudgetGPU #gpu #LinusTorvalds #Linux #PCSetup
https://blazetrends.com/linux-creator-linus-torvalds-setup-reveals-surprising-budget-amd-rx-580-gpu/?fsp_sid=82116

#GodotEngine rendering a spinning building using the #GPU in #Linux on a #Quest3 #XR headset