LLM Inspector, a new open-source CLI tool from developer Helal Saoudi, analyzes live LLM inference processes to show exactly how GPU memory is used by weights, KV cache, and workspace, then projects o…
NVIDIA and Hugging Face have integrated the NVIDIA NeMo Automodel open-source library with the Hugging Face Diffusers library, enabling production-grade distributed fine-tuning of video and image diff…
A new benchmark, Atrex-Bench, drawn from full-cluster production inference traces shows that even the best frontier coding agents reach only ~10% of hardware roofline on real GPU operators, with corre…
LingBot-Map, a feed-forward 3D foundation model for streaming scene reconstruction, achieves state-of-the-art performance at approximately 20 FPS on 518×378 resolution over sequences exceeding 10,000 …
A new book, 'Time Series with PyTorch' by Graeme Davidson and Lei Ma, offers a practical guide for time-series forecasting using PyTorch, covering evaluation, architectures, and tasks beyond point for…
A former pilot turned entrepreneur founded RendeRex, a Dubai-based hardware company, in 2020 during the COVID-19 pandemic, initially aiming to sell 3D rendering workstations but pivoting to AI worksta…
The Little Book of Reinforcement Learning, a short introduction to RL from basics to applied algorithms, is available on GitHub with PyTorch implementations and supplementary proofs. The book, authore…
Trellis2.c, a native executable for local 3D generation with Vulkan and CUDA backends, has been released on GitHub. The project aims to provide a lightweight alternative to Python/PyTorch runtimes, si…
A new arXiv paper (2607.13042v1) traces how PyTorch's automatic differentiation engine computes gradients for Physics-Informed Neural Network training, using a 1-3-3-1 multilayer perceptron and the in…
An AI-generated GPU kernel ran 18.71× faster than an optimized PyTorch baseline, with Fable 5 outperforming other models like Claude Opus 4.8 and GPT-5.5. The achievement raises questions about code r…
A developer released Flippermind-lite, a Super Tiny Language Model (STLM) designed to run on small devices like the Flipper Zero and Raspberry Pi. The project uses Qwen2.5-0.5B and includes fixes for …
A developer demonstrated running Google's Gemma 4 26B large language model on a 13-year-old Intel Xeon E5 v2 processor using CPU-only optimization techniques like 4-bit quantization and memory-efficie…
PyTorch 2.13, released this week, brings FlexAttention support for Apple Silicon with up to 12x speedup on sparse patterns and a new fused LinearCrossEntropyLoss operator that cuts peak GPU memory by …
MarkTechPost published a tutorial on building a Gin Config controlled PyTorch pipeline that separates training code from experiment configuration using .gin files. The pipeline implements a configurab…
The PyTorch-Triton 3.7 release introduces the Triton Plugin Extensions system, a framework for dynamically loading custom compiler passes, dialects, and DSL extensions into upstream Triton at runtime …
Unsloth AI Founding Engineer Challenge #1 produced a Triton GPU kernel for NF4 dequantization that achieves 1.27x–1.72x speedup over the existing bitsandbytes C++ implementation across all tested tens…
An architect at an unnamed company encountered an unexpected challenge when deploying a Visual Element Detection (VED) service on Azure Kubernetes Service (AKS): the target region did not offer the NV…
Vocalinux 0.14 beta has been released for offline voice dictation on Linux, adding configurable keyboard shortcuts, FunASR/SenseVoice remote API support, Wayland fixes, and improved CPU-based Whisper.…
A developer ported Karpathy's nanochat to run on a TPU v6e-8 using JAX, achieving a CORE score that reproduces the original quality but with a model FLOPs utilization (MFU) of about 24%, roughly half …
George Hotz's tinygrad claims to map PCIe Base Address Registers directly for NVIDIA and AMD GPUs, a low-level runtime approach that bypasses vendor driver APIs like CUDA and ROCm. The tinygrad runtim…