#directml kết quả tìm kiếm
spécifique jusque là inaccessible mais en software mobilisent une partie des 12,1 teraflops de calcul. l'API #DirectX12Ultimate déverrouille l'accès au Dual compute Unit pas moin de 13 teraflops qui sera utilisé pour le RT mais aussi #DirectML le machine learning IA. 👀⬇️
We've known for a long time that #DirectML hardware was announced at the time of the release of the #XboxSeries - just look at the Apu diagram !
🚀 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: gpuopen.com/learn/onnx-dir…
Microsoft hyped Xbox Series X | S DirectML before launch. Years later, nothing has come from it🔧 Digital Foundry says it's likely a dead end 🚨Sony didn't market Machine Learning or give it a name - Yet PS5 has delivered more meaningful ML Upgrades so far #Xbox #PS5 #DirectML
#DirectML at Build 2023 #MSBuild elevenforum.com/t/directml-at-…
@Microsoft will use Dual Compute Units to manage AI (#DirectML) in addition to raytracing (#DXR). Don't forget that @Microsoft's custom full RDNA2 APU for the #XboxSeries was designed in collaboration with @Amd engineers. 👀⬇️
⬛ 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: gpuopen.com/learn/radeon_g… (🧵 4/7)
#DirectML で遊んでいて、Windowsの場合はPythonが3.7でも動作するけど、WSL2だとPython 3.8ならOK、3.7だとNGってちょっと罠。 なんとなく3.7で環境作ろうとして何度かやって失敗したから試しに3.8にしたら動作した。 learn.microsoft.com/ja-jp/windows/…
🛠️Debugging AI on DirectML: Encountered a `uint8_t overflow` error while running @huggingface GPT-Neo 1.3B on AMD Radeon RX 6700 XT using `torch-directml`. This issue is specific to the `masked_fill` operation in the causal mask creation. #AI #MachineLearning #DirectML #AMD
Yes, because the #XboxSeries have dual compute units that can be used for #DirectML (AI)! The #FSR4 is software coupled with AI like the #PSSR ! 👀⬇️
Announcing preview support for #Llama2 in #DirectML elevenforum.com/t/announcing-p…
Xbox has officially opened the generative AI floodgates, and Phil Spencer wants to use it for game preservation and save titles that "were maybe tied to unique pieces of hardware" #DirectML #AutoSR #Xbox #Microsoft ☑️ gamesradar.com/games/xbox-has…
4K/60fps et RT hardware (DXR) sur #ForzaMotorsport . J'espère que les éditeurs tiers utiliseront les spécificités de chaque consoles ?! Les features : #VRS dans sa version 2.0 actuellement, #DirectML,#DXR, #DirectStorage,#VelocityArchitecture etc... 👀⬇️
Playing with @AMDIndonesia @AMD Ryzen 5 Pro 4650G with Vega 7 APU. iGP GFX allocation memory set to 16GB VRAM. running everyjourney checkpoint with 512x512, generated #aigirls result in just ~1 minute. #democratizeAI #directML
#DirectML を試したくてAMDなマシンを買ったので、ひとまず動きそうなところまで確認。 どのくらいサクサク動くものなのか分からないけど、NVIDIAのボードとかCUDAなしでもPyTorchで遊べそう。 github.com/microsoft/Dire…
🚀 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: gpuopen.com/learn/onnx-dir…
インストール済のWindows 11 + WSL:Ubuntu 24.04.1 LTS 上で動作する"公式ROCm 6.4.1にはgfx1151が無い"ため,ROCmは断念.動作確認済の #ComfyUI を #EVOX2 with #DirectML で動かすをシェア😌 (続く) ・Windows DirectML:Strix Halo GPU用ComfyUIセットアップガイド（最短版） claude.ai/public/artifac…
🛠️Debugging AI on DirectML: Encountered a `uint8_t overflow` error while running @huggingface GPT-Neo 1.3B on AMD Radeon RX 6700 XT using `torch-directml`. This issue is specific to the `masked_fill` operation in the causal mask creation. #AI #MachineLearning #DirectML #AMD
はてなブログに投稿しました Chromadbのテキスト埋め込み関数でGPUを使う - AOI-CAT's diary aoicat.hatenablog.com/entry/2025/05/… #はてなブログ #Chroma #DirectML
aoicat.hatenablog.com
Chromadbのテキスト埋め込み関数でGPUを使う - AOI-CAT's diary
はじめに Chromaのデフォルトテキスト埋め込み関数（ONNXMiniLM_L6_V2）は、GPUの使用をサポートしている。 nVidiaのGPU（cuda環境）ならonnxruntime-gpuをインストールして、関数の引数 preferred_providers にGPUをサポートするプロバイダーを設定する。 今…
⬛ 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: gpuopen.com/learn/radeon_g… (🧵 4/7)
Xbox has officially opened the generative AI floodgates, and Phil Spencer wants to use it for game preservation and save titles that "were maybe tied to unique pieces of hardware" #DirectML #AutoSR #Xbox #Microsoft ☑️ gamesradar.com/games/xbox-has…
Yes, because the #XboxSeries have dual compute units that can be used for #DirectML (AI)! The #FSR4 is software coupled with AI like the #PSSR ! 👀⬇️
#FSR4 utilisant l'IA ?! Sachant que seul le #XboxSeries ont l'IA hardware #DirectML (voir le diagramme du socle en bas) et qu'elles soient pas retenue par les studios tiers qui n'utiliserai pas le #FSR dans leurs jeux les jeux sur #XS pourraient utiliser le #FSR4. ⬇️
The #XboxSeries have #DXR for hardware raytracing and the same goes for #DirectML AI! It's official and it's not because some studios aren't using it for reasons we'll probably never know, but I have my own idea that ⬇️
We've known for a long time that #DirectML hardware was announced at the time of the release of the #XboxSeries - just look at the Apu diagram !
Exciting to see the #ONNX and #DirectML ecosystem in action! At #Caduceus, we support cutting-edge AI applications with decentralized edge rendering, perfect for #AI and #AIGC innovations. Whether it's AI, DePIN, or blockchain solutions, we’re ready to empower your journey.…
Windowsの機械学習向けAPI「DirectML」が「Copilot+ PC」に対応、WebNNも利用可能に #ITニュース #DirectML #CopilotPC #SnapdragonX #CodeZine dlvr.it/TCtdHs
codezine.jp
Windowsの機械学習向けAPI「DirectML」が「Copilot+ PC」に対応、WebNNも利用可能に
米Microsoftは、Windowsが備える機械学習向けAPI「DirectML」が、米Qualcommの「Snapdragon X」シリーズに対応したと8月29日（現地時間）に発表した。DirectMLはWindows PCが備えるGPU（Graphics Processing Unit）の機能を機械学習向けに抽象化したAPI。これまでは米AMDや米Intel、米NVIDIAのGPUの...
Windowsの機械学習向けAPI「DirectML」が「Copilot+ PC」に対応、WebNNも利用可能に #DirectML #CopilotPC #SnapdragonX #Adreno #ITニュース dlvr.it/TCtd21
it.f-frontier.com
Windowsの機械学習向けAPI「DirectML」が「Copilot+ PC」に対応、WebNNも利用可能に | ITちゃんねる
米Microsoftは、Windowsが備える機械学習向けAPI「DirectML」が、米Qualcommの「Snapdragon X」シリーズに対応したと8月29日（現地時間）に発表した。
Nieuws: De lancering van Windows Copilot+pc’s ligt al enkele maanden achter ons. Ondersteuning voor developers is nog niet wat het moet zijn, maar Microsoft beweegt in de goede richting… dlvr.it/TCYxlQ #Devops #Copilot #DirectML #QualcommSnapdragonXElite #WindowsonARM
techzine.nl
Microsoft zet broodnodige stap: DirectML draait op Copilot+-pc's
DirectML-support is een broodnodige feature voor Snapdragon X Elite-chips. Windows op Arm kan weer wat beter taken afronden.
DirectML support is a much-needed feature for Snapdragon X Elite chips. Windows on Arm can complete tasks a little better again. dlvr.it/TCYTYV #Devops #Copilot #DirectML - Follow for more
techzine.eu
Microsoft adds critical feature: native DirectML for Copilot+ PCs
The June launch of Windows Copilot+ PCs has come and gone. Support for developers is not yet at the level it should be, but Microsoft is moving in the
DirectML-support is een broodnodige feature voor Snapdragon X Elite-chips. Windows op Arm kan weer wat beter taken afronden. dlvr.it/TCYR6l #Devops #Copilot #DirectML #QualcommSnapdragonXElite #WindowsonARM
techzine.nl
Microsoft zet broodnodige stap: DirectML draait op Copilot+-pc's
DirectML-support is een broodnodige feature voor Snapdragon X Elite-chips. Windows op Arm kan weer wat beter taken afronden.
spécifique jusque là inaccessible mais en software mobilisent une partie des 12,1 teraflops de calcul. l'API #DirectX12Ultimate déverrouille l'accès au Dual compute Unit pas moin de 13 teraflops qui sera utilisé pour le RT mais aussi #DirectML le machine learning IA. 👀⬇️
@Microsoft will use Dual Compute Units to manage AI (#DirectML) in addition to raytracing (#DXR). Don't forget that @Microsoft's custom full RDNA2 APU for the #XboxSeries was designed in collaboration with @Amd engineers. 👀⬇️
Yes, because the #XboxSeries have dual compute units that can be used for #DirectML (AI)! The #FSR4 is software coupled with AI like the #PSSR ! 👀⬇️
We've known for a long time that #DirectML hardware was announced at the time of the release of the #XboxSeries - just look at the Apu diagram !
4K/60fps et RT hardware (DXR) sur #ForzaMotorsport . J'espère que les éditeurs tiers utiliseront les spécificités de chaque consoles ?! Les features : #VRS dans sa version 2.0 actuellement, #DirectML,#DXR, #DirectStorage,#VelocityArchitecture etc... 👀⬇️
Playing with @AMDIndonesia @AMD Ryzen 5 Pro 4650G with Vega 7 APU. iGP GFX allocation memory set to 16GB VRAM. running everyjourney checkpoint with 512x512, generated #aigirls result in just ~1 minute. #democratizeAI #directML
#DirectML を試したくてAMDなマシンを買ったので、ひとまず動きそうなところまで確認。 どのくらいサクサク動くものなのか分からないけど、NVIDIAのボードとかCUDAなしでもPyTorchで遊べそう。 github.com/microsoft/Dire…
#DirectML で遊んでいて、Windowsの場合はPythonが3.7でも動作するけど、WSL2だとPython 3.8ならOK、3.7だとNGってちょっと罠。 なんとなく3.7で環境作ろうとして何度かやって失敗したから試しに3.8にしたら動作した。 learn.microsoft.com/ja-jp/windows/…
🛠️Debugging AI on DirectML: Encountered a `uint8_t overflow` error while running @huggingface GPT-Neo 1.3B on AMD Radeon RX 6700 XT using `torch-directml`. This issue is specific to the `masked_fill` operation in the causal mask creation. #AI #MachineLearning #DirectML #AMD
Voilà pourquoi la @Microsoft n'a pas besoin de #XboxSeriesXPro! Je rappelle que l'apu de la #XboxSeries a du #DirectML en hardware pas la #PS5 !
Latest DirectML boosts AMD GPU AWQ-based LM acceleration Read more on govindhtech.com/latest-directm… #DirectML #amd #gpu #llm #microsoft #awq #ai #amdradeon #amdryzen #technology #technews #news #govindhtech @AMD @TechGovind70399
Learn how to increase #ONNXRuntime and #DirectML performance to make the most of your NVIDIA hardware. Get a technical walkthrough, including example code, in part 6 of our optimizing end-to-end #AI blog series: bit.ly/442tT8a
Something went wrong.
Something went wrong.
United States Trends
- 1. #UFC322 198K posts
- 2. Islam 308K posts
- 3. #LingTaoHeungAnniversary 703K posts
- 4. LING BA TAO HEUNG 704K posts
- 5. Good Sunday 54.6K posts
- 6. Morales 39.5K posts
- 7. Wuhan 12.6K posts
- 8. Ilia 9,065 posts
- 9. #GirlPower N/A
- 10. Prates 38.3K posts
- 11. Valentina 16.8K posts
- 12. #INDvsSA 19.8K posts
- 13. Khabib 16.5K posts
- 14. Dagestan 4,479 posts
- 15. Sark 6,439 posts
- 16. #ONE173 16.8K posts
- 17. Shevchenko 13.7K posts
- 18. Dillon Danis 16.6K posts
- 19. Ole Miss 12.9K posts
- 20. Topuria 7,169 posts