TGINSIGHT SIMILAR POSTS
Source channel @githubtrending · Post #15428 · Jan 22
#cplusplus FlashMLA is DeepSeek's optimized attention library that makes AI models run faster and use less memory. It works with advanced NVIDIA GPUs to speed up how language models process information, achieving up to 660 trillion floating-point operations per second. The library supports both dense and sparse attention modes, meaning it can focus on important tokens while skipping less relevant ones, reducing computational waste. For you, this means faster AI responses, lower costs for running large language models, and better performance on tasks like chatbots and code generation. The technology is open-source and integrates with popular AI frameworks like PyTorch and Hugging Face, making it accessible for developers building next-generation AI applications. https://github.com/deepseek-ai/FlashMLA
Hashtags
Search: #dl
@repo_science · Post #4131 · 05/18/2024, 09:06 PM
#DL 📱 Zeus New Pytorch Ecosystem Tool Zeus is an open source toolkit for measuring and optimizing power consumption of deep learning workloads. 🖥Github ----- Main channel: @repo_science Coupons: @freecoupons_reposcience -----
Hashtags
@amneumarkt · Post #706 · 11/21/2025, 07:53 AM
#dl Introducing more symmetries in attention https://github.com/NVIDIA/torch-harmonics https://neurips.cc/virtual/2025/loc/san-diego/poster/117783
Hashtags
@amneumarkt · Post #691 · 10/05/2025, 07:41 AM
#dl Park, Chanwook, Sourav Saha, Jiachen Guo, Hantao Zhang, Xiaoyu Xie, Miguel A. Bessa, Dong Qian, et al. 2025. “Unifying Machine Learning and Interpolation Theory via Interpolating Neural Networks.” Nature Communications 16 (1): 1–12. https://www.nature.com/articles/s41467-025-63790-8
Hashtags
@amneumarkt · Post #683 · 06/28/2025, 07:04 AM
#dl A few cool ideas in this model. Introducing Gemma 3n: The developer guide - Google Developers Blog https://developers.googleblog.com/en/introducing-gemma-3n-developer-guide/
Hashtags
@amneumarkt · Post #681 · 06/14/2025, 08:43 AM
#dl So tensorflow and jax are deprecated in the transformers package. https://github.com/huggingface/transformers/pull/38758
Hashtags
@amneumarkt · Post #662 · 02/21/2025, 07:42 AM
#dl Sure, nobody uses it anyways.
Hashtags
@amneumarkt · Post #625 · 10/03/2024, 09:31 PM
#dl PyTorch Native Architecture Optimization: torchao | PyTorch https://pytorch.org/blog/pytorch-native-architecture-optimization/
Hashtags
@amneumarkt · Post #624 · 10/03/2024, 06:07 PM
#dl https://soumith.ch/blog/2024-10-02-training-10k-scale.md.html
Hashtags
@amneumarkt · Post #602 · 07/20/2024, 05:48 AM
#dl There is this new lib called scale. One could compile CUDA code to use it on AMD GPU. https://docs.scale-lang.com/manual/how-to-use/ I don't know who is more pissed off, NVidia or AMD.
Hashtags
@amneumarkt · Post #556 · 03/16/2024, 09:09 AM
#dl This repo is really nice. yuanchenyang/smalldiffusion: Simple and readable code for training and sampling from diffusion models https://github.com/yuanchenyang/smalldiffusion
Hashtags
@amneumarkt · Post #506 · 11/13/2023, 08:30 AM
#dl Google & USC benchmarked a prompt based forecasting method, and the results are amazing. Cao D, Jia F, Arik SO, Pfister T, Zheng Y, Ye W, et al. TEMPO: Prompt-based Generative Pre-trained Transformer for time series forecasting. arXiv [cs.LG]. 2023. Available: http://arxiv.org/abs/2310.04948
Hashtags
@amneumarkt · Post #484 · 06/12/2023, 06:13 AM
#dl https://twitter.com/armin_kekic/status/1667181047480479751?s=20
Hashtags