Google Developers BlogGoogle's AI Edge team has open-sourced ML Drift, a high-performance, cross-platform edge device GPU computing engine for on-device AI/ML inference. By abstracting the complexity of hardware low-level APIs such as OpenGL ES, OpenCL, Metal, and WebGPU, ML Drift enables developers to build real-time interactive AI experiences.
PyTorch BlogThe EUROPA Union successfully trained a MoE model with 951.9B parameters on the JUPITER supercomputer using PyTorch and TorchTitan, increasing single GPU throughput from 90 to approximately 1,300 tokens/s with 384 NVIDIA GH200 GPUs.
vLLM BlogFeatured7272NVIDIA Research and Inferact's REINFORCE Zero proposes to first align rollout with training kernel, then use unclipped token-level importance sampling to correct weight lag, achieving a training step throughput 2.40 times that of synchronous training under 16 GPUs and 12K updates.
Why it matters: The paper breaks down the gap between asynchronous RL's rollout and training into two items: weight lag and kernel inconsistency, and proposes an alignment-free, no-trimming importance sampling method after alignment.
IBM Research Blog3636ORNL, IBM, and AMD deploy a quantum error correction (QEC) collaborative design platform at IBM's New York Poughkeepsie campus, integrating the IBM Quantum Nighthawk r2 processor with the ORNL Umbra HPC system, which includes the AMD Instinct MI355X GPU and 5th Gen AMD EPYC™ CPU, providing low-latency quantum-classical execution capabilities for algorithm development and validation in hybrid HPC-quantum computing environments.
NVIDIA BlogC-BRAIN and NVIDIA are using AI to change the pace of dementia research. They are accelerating the development of new therapies by integrating previously unpublished Alzheimer's research data, including unpublished results and controlled proprietary data. C-BRAIN employs a hybrid centralized and federated architecture to protect data privacy, deploying 64 NVIDIA data center GPUs and cloud nodes, and plans to introduce the NVIDIA Blackwell system to support its expansion goals.
vLLM BlogFeatured6060The vLLM team used P/D separation and Wide-EP to serve DeepSeek-V4.1-Flash on a single NVIDIA GB300 NVL72 rack, achieving 954K tokens/s per GPU and 196 tokens/s for single-user P90 on SemiAnalysis AgentX, which is 7.9 times the early single-node results.
Why it matters: The vLLM team has released per-item optimization and measured data for serving DeepSeek-V4.1-Flash on GB300 NVL72, which can be compared with DeepSeek's own engine DeepInfer.
Databricks BlogDatabricks AI Runtime (AIR) has been officially released, supporting developers to train and fine-tune open models on serverless NVIDIA GPU using NVIDIA NeMo, NGC containers, and Nemotron new recipes.
NVIDIA Blog5555GeForce NOW will introduce 'Warhammer 40,000: Dawn of War IV', supporting cloud gaming experience of the Battle of Cronus; DLSS 5 technology will launch in 'War Thunder' and 'Where Winds Meet', performance improved tenfold over the previous version; 10 new games added this week, including 'Dragon's Dogma 2: Awakening of the Dark Elves'
Hugging Face BlogFeaturedSame story · 2 outlets8585Mistral releases multimodal MoE model Mistral Large 4 (Le Chonk), with approximately 1T total parameters, 49B activation parameters, a context window of 1M tokens, offering two official versions FP8 and NVFP4, as well as GGUF made by Unsloth and llama.cpp teams.
Why it matters: The original text provides specific scores for the 1T total parameter, 49B activation multimodal MoE model on cybersecurity and coding benchmarks, along with measured throughput for local and endpoint deployment.
Meta EngineeringMeta Andromeda's Next-Generation Ad Retrieval Engine Achieves Near-Ranking-Level Model Performance at the Retrieval Layer, Processing Over 1 Billion Ads Per Card Per Second, System Throughput Reaches 100 Billion Times Per Second, 100 Times Higher Than the 2024 Version, and Brings a 2.4% Conversion Rate Increase for Advertisers.
NVIDIA BlogNVIDIA showcased its full lineup of local AI hardware at GTC Berlin, with RTX Spark now available to support personal agents on Windows PCs, DGX Spark enabling developers to build and scale Linux applications, and DGX Station designed for Linux enterprise teams to run cutting-edge workloads, with the Windows version launching in December.
Hugging Face BlogFeatured6262The Hugging Face PyTorch team introduces updates to the MPS backend in PyTorch 2.15, adding the tensor torch.float8_e4m3fn and F.scaled_mm, with FP8 tensors occupying half the space of BF16.
Why it matters: The original text provides specific numbers for the MPS backend in FP8, MoE decoding, and memory usage, which can be used to judge the actual changes in running models locally on Mac.
Google DeepMindGemini 4 Argon has achieved major breakthroughs in physical understanding, code generation, and visual semantic reasoning, achieving a pure reasoning score of 87% on the Embodied Reasoning (ERQA) benchmark test, surpassing all leading models. Argon can be applied to all stages of the robot's lifecycle, including mobile manipulation practice using MuJoCo, zero-shot robot control, and closed-loop steering of video world models.
NVIDIA BlogNVIDIA RTX Spark integrates the capabilities of gaming PCs, creative studios, and personal AI supercomputers into a single superchip, equipped with up to 128GB of unified memory and 1 petaflop of AI computing power.
Google Developers BlogGemini 4 Argon has made significant breakthroughs in physical understanding, code generation, and visual semantic reasoning, achieving an 87% score on the Embodied Reasoning (ERQA) benchmark test, surpassing all state-of-the-art models.
NVIDIA BlogNVIDIA and Microsoft collaborate to launch RTX Spark laptops and Windows AI agent infrastructure, supporting local running of 125B parameter Qwen 3.8 Flash Next models; Microsoft releases MXC operating system-level container to ensure agent secure operation, pre-orders for RTX Spark open today, paired with Grace Blackwell chip, offering up to 20 PFLOPS inference performance.
Google Developers BlogGoogle DeepMind released EmbeddingGemma 2, an open-weight multimodal embedding model with only 740M parameters, capable of unifying text, images, video frames, and audio into a single vector space.
Microsoft ResearchGigaPath-Flash and GigaTIME-Flash significantly reduce computational requirements through streamlined ViT-S encoders, making large-scale pathological research more feasible while maintaining performance.
