Huge thanks to our compute providers @CoreWeave and the EuroHPC, @tri_dao and @d_haziza for their help with FlashAttention and xFormers, and @huggingface
, vLLM, @skypilot_org
, FastChat for their help and support with this release.
CODE
-
Thanks to compute providers and partners for model release support
By
–
-
GQA and Sliding Window Optimization for Efficient Decoding
By
–
We trained it with GQA and a sliding window of 4096 tokens, resulting in constant cache size and a linear decoding speed. Our changes to FlashAttention v2 and xFormers to support sliding window are available to the community.
-
Mistral AI team builds advanced MLOps stack in three months
By
–
Very proud of the Mistral AI team who rebuilt a top-performance MLops stack, and designed a very sophisticated data processing pipeline, from scratch, in less than 3 months.
-
Cloudflare Launches Workers AI for Accessible Developer Inference
By
–
What's going on here?
Cloudflare is launching Workers AI to make AI inference easy and accessible for all developers. What does this mean? Workers AI allows developers to run popular AI models with just a few lines of code, without needing to manage infrastructure. It is -
Sanjay Ghemawat: Long-Time Colleague and AI Collaborator
By
–
My long time colleague and collaborator, Sanjay Ghemawat.
-
Mistral 7B v0.1 Model Release Available for Download
By
–
magnet:?xt=urn:btih:208b101a0f51514ecf285885a8b0f6fb1a1e4d7d&dn=mistral-7B-v0.1&tr=udp%
http://
2Ftracker.opentrackr.org1337announce&tr=https%https://t.co/HAadNvH1t0443announce RELEASE ab979f50d7d406ab8d0b07d09806c72c -
Self-Attention Mechanism in Large Language Models Explained
By
–
Understanding and Coding the Self-Attention Mechanism of Large Language Models From Scratch https://
bit.ly/3YL821l #AI #DeepLearning #MachineLearning #DataScience -
Training YOLOv8 Object Detection with KerasCV and TensorFlow
By
–
🚀Blog-tastic Tuesdays🚀https://t.co/qDpaFplJU0
— Satya Mallick (@LearnOpenCV) 26 septembre 2023
YOLOv8 is the latest addition to the KerasCV library. With an easy training pipeline and high performance, it is now a breeze to use YOLOv8 with TensorFlow and Keras. Our latest article shows how to train the KerasCV YOLOv8 model… pic.twitter.com/WDV5JuiKaXBlog-tastic Tuesdays https://
learnopencv.com/object-detecti
on-using-kerascv-yolov8/
… YOLOv8 is the latest addition to the KerasCV library. With an easy training pipeline and high performance, it is now a breeze to use YOLOv8 with TensorFlow and Keras. Our latest article shows how to train the KerasCV YOLOv8 model -
What’s New in R2023b Release Updates
By
–
What's New in R2023b? ➡️ https://t.co/85CDdswX6d pic.twitter.com/jBNii8zvQB
— MathWorks (@MathWorks) 26 septembre 2023What's New in R2023b? http://
spr.ly/6017u6L6f -
Ludwig v0.8.4 Release: Major LLM Enhancements and Improvements
By
–
We're pleased to introduce an exciting new minor release (v0.8.4) that brings significant enhancements to your @ludwig_ai experience including enhancements for #LLMs. Thank you to all of our amazing contributors! Download Ludwig v0.84: https://
pbase.ai/3rqHiZg.