Machine learning news — deep learning, reinforcement learning, neural architecture search, diffusion models, and new ML frameworks and libraries.
Helion kernels were integrated into vLLM for FP8 inference using Qwen3 models and evaluated across NVIDIA H100 and B200 GPUs. The experiments demonstrated that Helion provides a productive PyTorch-native workflow for dev…
Python Software Foundation security developer Seth Larson reported that JetBrains' PyCharm IDE Full Line Code Completion plugin suggests code containing severe vulnerabilities, but JetBrains declined to classify the defe…
PALO ALTO, Calif. — Startup Ricursive, founded by the former leads of Google's AlphaChip project, announced plans to develop an end-to-end AI model for chip design, aiming to transform the industry by enabling custom chi…
NVIDIA has optimized Google DeepMind's new DiffusionGemma model to run up to four times faster on its GeForce RTX GPUs, RTX PRO workstations, and DGX Spark systems. Unlike traditional language models that generate text o…
Google DeepMind released DiffusionGemma, an experimental open-source text diffusion model that generates text up to four times faster than traditional large language models by processing entire blocks of text simultaneou…
AWS announced Neuron Agentic Development capabilities, a collection of AI agents and skills that automate kernel optimization for AWS Trainium and Inferentia hardware. The tools enable developers to write, debug, and pro…
Amazon Web Services has released a solution for building an AI-powered equipment repair assistant using Amazon Bedrock AgentCore that helps farmers and field technicians diagnose heavy machinery problems, identify requir…
The "Memory Wall" bottleneck, originally coined in the 1990s to describe the speed gap between processors and DRAM, has re-emerged in the AI era as DRAM and high-bandwidth memory (HBM) struggle to meet the massive storag…
Adi Polak, a distributed systems expert and author, presented a framework for moving AI systems beyond simple prompting to context engineering and memory management at scale during a recent talk. Polak argued that effect…
A research group at the University of Twente in the Netherlands has demonstrated a method to reduce energy consumption during large language model training by up to 14 percent without sacrificing speed. The technique, ca…
Python developers can use virtual environments to isolate project dependencies and avoid conflicts between incompatible package versions. The native `venv` tool, included in all supported Python versions, creates separat…
Pyrefly, a Python type checker, has introduced an experimental feature that brings tensor shapes into Python's type system, allowing shape annotations to become inferred type hints instead of comments. The feature was pr…
At Anyscale's Ray Day: NYC event, Torc Robotics reported achieving 90% GPU utilization by consolidating its fragmented multimodal AI stack onto Ray, up from 30-40%. Discord detailed its ML platform evolution from open-so…
A high-performance Expert Parallelism (EP) kernel is essential for running large Mixture-of-Experts (MoE) language models across multiple GPUs, as it handles the dynamic routing of tokens to experts located on different …
NVIDIA Isaac Lab on Amazon SageMaker AI now enables robotics teams to train reinforcement learning policies for humanoid robots at scale, using either SageMaker HyperPod for persistent cluster training or SageMaker Train…
Google announced Gemini 3.5 Live Translate, a speech-to-speech AI model that provides instant voice-to-voice translation in over 70 languages. The model, part of the Gemini 3.5 family launched at I/O, processes speech co…
NVIDIA released a workflow for converting FP8-quantized CLIP model checkpoints into TensorRT engines, enabling faster inference and higher GPU throughput for production deployment. The process involves exporting the quan…
Google released Gemini 3.5 Live Translate, a streaming speech-to-speech audio model that translates over 70 languages in real time. The model processes audio continuously rather than waiting for a speaker to finish, pres…
NVIDIA introduced an automated, AI-driven research loop called Auto-FL within its NVIDIA FLARE platform to help researchers test and optimize federated learning strategies more efficiently. The system uses bounded AI age…
Google Cloud's GKE Inference Gateway delivers up to 92.8% shorter wait times and 62.6% lower inter-token latency compared to the next leading managed Kubernetes service, according to an independent benchmark report. The …