#directml — Public Fediverse posts
Live and recent posts from across the Fediverse tagged #directml, aggregated by home.social.
-
すべてのPCに「Xbox モード」を展開 ～「GDC 2026」に先駆けMicrosoftが発表／過去10年で最大のツール拡充、初回ロード時間の大幅短縮や「DirectStorage」の強化も
https://forest.watch.impress.co.jp/docs/news/2092626.html#forest_watch_impress #DirectX #DirectStorage #DirectML #Xbox_モード #Advanced_Shader_Delivery #ASD #XBox_FSE #エンタメ #Windows #プログラミング
-
🚀 Unlock blazing-fast neural network inference on AMD hardware!
Discover how to maximize #ONNX Runtime + #DirectML with DirectX 12 for seamless GPU pipelines.
🛠️ Dive into practical setup tips and code samples in our latest guide: https://gpuopen.com/learn/onnx-directlml-execution-provider-guide-part1/?utm_source=mastodon&utm_medium=social&utm_campaign=directml
-
🚀 Unlock blazing-fast neural network inference on AMD hardware!
Discover how to maximize #ONNX Runtime + #DirectML with DirectX 12 for seamless GPU pipelines.
🛠️ Dive into practical setup tips and code samples in our latest guide: https://gpuopen.com/learn/onnx-directlml-execution-provider-guide-part1/?utm_source=mastodon&utm_medium=social&utm_campaign=directml
-
⬛ No more black boxes! AMD Radeon GPU Profiler (RGP) now supports AMD RDNA 4 GPUs.
🔥 Profiling of #DirectML apps and compute-based #DX12 and #Vulkan apps
🔥 Improved support for Work Graphs apps
🔥 UI enhancements to the ISA views
More info: https://gpuopen.com/learn/radeon_gpu_profiler_2-4_support_rx9000_series_pure-compute_applications/?utm_source=mastodon&utm_medium=social&utm_campaign=rdts
(🧵 4/7) -
⬛ No more black boxes! AMD Radeon GPU Profiler (RGP) now supports AMD RDNA 4 GPUs.
🔥 Profiling of #DirectML apps and compute-based #DX12 and #Vulkan apps
🔥 Improved support for Work Graphs apps
🔥 UI enhancements to the ISA views
More info: https://gpuopen.com/learn/radeon_gpu_profiler_2-4_support_rx9000_series_pure-compute_applications/?utm_source=mastodon&utm_medium=social&utm_campaign=rdts
(🧵 4/7) -
DirectML unlocks new silicon for AI experiences across Windows Copilot+ PCs.
https://blogs.windows.com/windowsdeveloper/2024/11/12/directml-unlocks-new-silicon-for-ai-experiences-across-windows-copilot-pcs/
#directml #ai #npu #windowscopilotruntime #copilotpluspc #msignite #windows #windows11 -
DirectML unlocks new silicon for AI experiences across Windows Copilot+ PCs.
https://blogs.windows.com/windowsdeveloper/2024/11/12/directml-unlocks-new-silicon-for-ai-experiences-across-windows-copilot-pcs/
#directml #ai #npu #windowscopilotruntime #copilotpluspc #msignite #windows #windows11 -
Save the Date: Windows Dev Chat - October 24, 2024.
#windowsdev #devcommunity #powertoys #terminal #ai #wpf #winui #directml #wsl
https://www.youtube.com/watch?v=5Gk3TDyv0IU&ab_channel=WindowsDeveloper -
Save the Date: Windows Dev Chat - October 24, 2024.
#windowsdev #devcommunity #powertoys #terminal #ai #wpf #winui #directml #wsl
https://www.youtube.com/watch?v=5Gk3TDyv0IU&ab_channel=WindowsDeveloper -
DirectML expands NPU support to Copilot+ PCs and WebNN.
#directml #ai #npu #copilot #webnn #CopilotPlusPC #onnx #windowsdev
https://blogs.windows.com/windowsdeveloper/2024/08/29/directml-expands-npu-support-to-copilot-pcs-and-webnn/ -
DirectML expands NPU support to Copilot+ PCs and WebNN.
#directml #ai #npu #copilot #webnn #CopilotPlusPC #onnx #windowsdev
https://blogs.windows.com/windowsdeveloper/2024/08/29/directml-expands-npu-support-to-copilot-pcs-and-webnn/ -
Домашняя нейронка. Какое «железо» выбрать для Stable Diffusion?
Нейросетями теперь пользуются почти все мои знакомые. Чат-боты с искусственным интеллектом в плане сбора информации намного удобнее поисковых систем, а нейросети, генерирующие картинки по запросу, нашли свое применение в рекламе и иллюстрировании контента. Однако как раз с ними не все столь гладко, как хотелось бы. Доступ к одним без плясок с бубном и нервотрепки невозможно оплатить из России, другие не радуют качеством изображений, путаясь в количестве пальцев, изгибая конечности персонажей невообразимым образом или выдавая в ответ на запрос совсем не то, что требовалось. Тут у меня и возникла идея развернуть Stable Diffusion на локальной машине, чтобы не мучиться с оплатой и токенами. А для этого сначала следует разобраться, какое железо потребуется для реализации подобной задумки.
https://habr.com/ru/companies/serverspace/articles/806639/
#Stable_Diffusion #SDXL #stability_ai #AUTOMATIC1111 #vladmandic #Fooocus #SHARK #nvidia #amd #DirectML #OpenVINO #Tensor #DiffusionBee
-
I can see this year Microsoft pushing DirectAI to enable ML/AI workloads into the hardware through DirectX.
It would be cool to let the game leverage these workloads without depending on a NVIDIA card.
It would push data to the cloud and receive results if the hardware doesn't have AI capabilities (or the result exceeds computational power).
Guess that would also translate into Xbox Sebille (2024), Ellewood and Brooklin (2025).
-
Introducing Neural Processor Unit (NPU) support in DirectML (developer preview).
#ai #machinelearning #windows #npu #directML
https://blogs.windows.com/windowsdeveloper/2024/02/01/introducing-neural-processor-unit-npu-support-in-directml-developer-preview/ -
Introducing Neural Processor Unit (NPU) support in DirectML (developer preview).
#ai #machinelearning #windows #npu #directML
https://blogs.windows.com/windowsdeveloper/2024/02/01/introducing-neural-processor-unit-npu-support-in-directml-developer-preview/ -
DirectML: Accelerating AI on Windows, now with NPUs.
#ai #windows #directML #intel #npu #hardware #machinelearning
https://blogs.windows.com/windowsdeveloper/2023/12/14/directml-accelerating-ai-on-windows-now-with-npus/ -
DirectML: Accelerating AI on Windows, now with NPUs.
#ai #windows #directML #intel #npu #hardware #machinelearning
https://blogs.windows.com/windowsdeveloper/2023/12/14/directml-accelerating-ai-on-windows-now-with-npus/ -
We're excited to reveal an average 2X performance improvement with #Microsoft Olive #DirectML Stable Diffusion 1.5 🔥
The just-released @AMD
#Adrenalin 23.5.2 driver includes optimizations for Microsoft Olive DirectML.⏩ Find out more on #GPUOpen:
-
We're excited to reveal an average 2X performance improvement with #Microsoft Olive #DirectML Stable Diffusion 1.5 🔥
The just-released @AMD
#Adrenalin 23.5.2 driver includes optimizations for Microsoft Olive DirectML.⏩ Find out more on #GPUOpen:
-
Die von Microsoft entwickelte Variante des Machine-Learning-Frameworks setzt zur Hardwarebeschleunigung auf die DirectX-API DirectML.
Machine Learning: TensorFlow-DirectML verlässt den Preview-Modus