@onnxruntime

Cross-platform training and inferencing accelerator for machine learning models.

onnxruntime.ai
Joined September 2018
  • Tweets282
  • Following42
  • Followers1.4K
45Media
ONNX Runtime & DirectML now support Phi-3 mini models cross-platforms & devices! Plus, the new ONNX Runtime Generate() API simplifies LLM integration into your apps. Try Phi-3 on your favorite hardware! Read more: onnxruntime.ai/blogs/acceler… #ONNX #DirectML #Phi3
3
14
2
28
4,782
Run PyTorch models in the browser, on mobile and desktop, with #onnxruntime, in your language and development environment of choice 🚀onnxruntime.ai/blogs/pytorch…
5
15
2,881
Developers, don't overlook the power of Swift Package Manager! It simplifies dependency management and promotes modularity. Plus, exciting news: ONNXRuntime just added support for SPM! #iOSdev #SwiftPM #ONNXRuntime
1
3
1,593
#ONNX Runtime saved the day with our interoperability and ability to run locally on-client and/or cloud! Our lightweight solution gave them the performance they needed with quantization & configuration tooling. Learn how they achieved this in this blog! cloudblogs.microsoft.com/ope…

ONNX Runtime: performant on-device inferencing | Microsoft Open Source Blog

The team at Pieces shares the problems and solutions evaluated for their on-device model serving stack and how ONNX Runtime enables their success.

opensource.microsoft.com
5
9
1,619
Give yourself a treat (like this adorable🐶 deserves) and read this blog on how to use #ONNX Runtime on #Android! devblogs.microsoft.com/surfa…
SurfaceDuoDev@surfaceduodev2 Feb 2023
Quick intro to @onnxruntime and applying #machinelearning on Android devblogs.microsoft.com/surfa…

On-device machine learning with ONNX - Surface Duo Blog

Hello Android developers, This week we’re going to get started with on-device machine learning using the ONNX Runtime and check out an Android sample that

devblogs.microsoft.com
1
979
📢 This new blog by @tryolabs is awesome! Learn how to fine-tune a NLP model and accelerate with #ONNXRuntime!
Tryolabs@tryolabs30 Jan 2023
Maximize the power of LLMs! 💬 Our step-by-step guide covers fine-tuning for specific NLP tasks w/ GPT-3, OPT, & T5. We shared everything from building custom datasets to optimizing inf time with @huggingface 🤗Optimum and @onnxai.🚀 bit.ly/3DqLXxb #LargeLanguageModels
2
4
958
Join us live TODAY! We will be talking to Akhila Vidiyala and Devang Aggarwal on AI Show with Cassie! We will show how developers can use #huggingface #optimum #Intel to quantize models and then use #OpenVINO for #ONNXRuntime to accelerate performance. 👇 aka.ms/aishowlive

AI Show LIVE | AI Agents for Beginners v2 & How to vibe code correctly

Tune in to The AI Show Live on Monday, October 27 where we discuss ...

youtube.com
1
1
6
977
In this blog, we will discuss how to make huge models like #BERT smaller and faster with #Intel #OpenVINO, Neural Networks Compression Framework (NNCF) and #ONNX Runtime through #Azure! 👇 cloudblogs.microsoft.com/ope…

OpenVINO™, ONNX Runtime, and Azure improve BERT inference speed | Microsoft Open Source Blog

Make large models smaller and faster with OpenVino Execution Provider, NNCF and ONNX Runtime leveraging Azure Machine Learning.

opensource.microsoft.com
2
5
634
👀
Jingya Huang@Jhuaplin24 Jan 2023
🚀 Want easier and faster training for your models on GPUs? Thanks to the @onnxruntime backend, 🤗 Optimum can help you achieve 39% - 130% acceleration with just a few lines of code change. Check out our benchmark results NOW! 👀 huggingface.co/blog/optimum-…

Optimum+ONNX Runtime - Easier, Faster training for your Hugging Face models

We’re on a journey to advance and democratize artificial intelligence through open source and open science.

huggingface.co
2
1
580
onnxruntime retweeted
ONNX@onnxai7 Dec 2022
We are seeking your input to shape the ONNX roadmap! Proposals are being collected until January 24, 2023 and will be discussed in February. Submit your ideas at forms.microsoft.com/pages/re…

Microsoft Forms

forms.cloud.microsoft
2
3
5
onnxruntime retweeted
Jingya Huang@Jhuaplin28 Nov 2022
Imagine the frustration of, after applying optimization tricks, finding that the data copying to GPU slows down your "MUST-BE-FAST" inference...🥵 🤗 Optimum v1.5.0 added @onnxruntime IOBinding support to reduce your memory footprint. 👀 github.com/huggingface/optim… More ⬇️

GitHub - huggingface/optimum: 🚀 Accelerate inference and training of 🤗 Transformers, Diffusers,...

🚀 Accelerate inference and training of 🤗 Transformers, Diffusers, TIMM and Sentence Transformers with easy to use hardware optimization tools - huggingface/optimum

github.com
1
15
54
onnxruntime retweeted
efxmarty@efxmarty27 Oct 2022
Want to use TensorRT as your inference engine for its speedups on GPU but don't want to go into the compilation hassle? We've got you covered with 🤗 Optimum! With one line, leverage TensorRT through @onnxruntime! Check out more at hf.co/docs/optimum/onnxrunti…
1
18
1
76
📣The new version of #ONNXRuntime v1.13.0 was just released!!! Check out the release note and video from the engineering team to learn more about what was in this release! 📝github.com/microsoft/onnxrun… 📽️youtu.be/vo9vlR-TRK4

Releases · microsoft/onnxruntime

ONNX Runtime: cross-platform, high performance ML inferencing and training accelerator - microsoft/onnxruntime

github.com
2
1
1
4
👀
ONNX@onnxai7 Oct 2022
Next up from #ONNXCommunityDay: Accelerating Machine Learning w/ @ONNXRuntime & @HuggingFace! In this session, @jeffboudier will show the latest solutions from #HuggingFace to deploy models at scale w/ great performance leveraging #ONNX & #ONNXRuntime. youtu.be/9H7biU4eLZY

Accelerating Machine Learning with ONNX Runtime and Hugging Face

Hugging Face has democratized state of the art machine learning wit...

youtube.com
1
3
Finally tokenization with Sentence Piece BPE now works as expected in #NodeJS #JavaScript with tokenizers library 🚀! Now getting "invalid expand shape" errors when passing text tokens' encoded ids to the MiniLM @onnxruntime converted @MSFTResearch model huggingface.co/microsoft/Mul…

microsoft/Multilingual-MiniLM-L12-H384 · ONNX Inference in NodeJS

I have successfully converted the sentencepiece.bpe.model to JSON vocabulary and text merges. See here for more details about BPE SentencePiece tokenizer. The tokenization works fine using tokenize...

huggingface.co
Loreto Parisi
@loretoparisi28 Sep 2022
Replying to @loretoparisi
Sentence piece vocabulary and merge files generated, some minor issues occurring, hopefully @huggingface can help 🙏github.com/huggingface/token…
5
3
🏭 The hardware optimization floodgates are open!🔥 Diffusers 0.3.0 supports an experimental ONNX exporter and pipeline for Stable Diffusion 🎨 To find out how to export your own checkpoint and run it with @onnxruntime, check the release notes: github.com/huggingface/diffu…
2
16
1
83
onnxruntime retweeted
Santosh@exendahal21 Sep 2022
Replying to @jfversluis
@jfversluis What about a video on ONNX runtime? Here is the official documentation devblogs.microsoft.com/xamar… And MAUI example: github.com/microsoft/onnxrun…

.NET MAUI

devblogs.microsoft.com
1
2
4
The natural language processing library Apache OpenNLP is now integrated with ONNX Runtime! Get the details and a tutorial explaining its use on the blog: msft.it/6013jfemt #OpenSource
A square made of smaller squares, circles, and people's faces.

ALT A square made of smaller squares, circles, and people's faces.

6
8
In this article, a community member used #ONNXRuntime to try out GPT-2 model which generates English sentences from Ruby language: dev.to/kojix2/text-generatio…

Text generation with GPT-2 in Ruby

Introduction Many people think that accessing AI from languages other than Python is too...

dev.to
1
3
2
8
Come join us for the hands on lab(September 28, 1-3pm)to learn about accelerating your ML models via ONNXRunTime frameworks on Intel CPUs and GPUs..some surprise goodies as well #IntelON #iamintel #intelarc @IntelGraphics @IntelSoftware @gfxlisa intel.com/content/www/us/en/… – at San Jose, CA
2
1
Load more