Nvidia Blackwell B200
PulseAugur coverage of Nvidia Blackwell B200 — every cluster mentioning Nvidia Blackwell B200 across labs, papers, and developer communities, ranked by signal.
10 day(s) with sentiment data
-
Microsoft hikes Xbox prices, introduces payment plan; SageMaker AI optimizes NVIDIA Blackwell training
Microsoft is raising the prices for its Xbox Series consoles and introducing a "Buy Now, Pay Later" option. Separately, a guide details how to optimize model training on Amazon SageMaker AI using NVIDIA Blackwell archit…
-
AWS SageMaker AI integrates NVIDIA Blackwell GPUs for optimized large model training
Amazon SageMaker AI is now optimized to leverage NVIDIA Blackwell GPUs, enabling more efficient training of large AI models. The new P6-B200 instances with 8 Blackwell GPUs offer expanded memory and higher bandwidth, re…
-
DFlash accelerates AI inference with parallel token block drafting · 2 sources tracked
Researchers from the University of California, San Diego, have developed DFlash, a novel speculative decoding technique that significantly accelerates AI inference. Unlike traditional methods that generate tokens one by…
-
NVIDIA releases quantized GLM-5.2 MoE model with 1M context
NVIDIA has released the GLM-5.2 NVFP4 model, a quantized version of ZAI's GLM-5.2. This Mixture-of-Experts model is optimized for reasoning and coding tasks, featuring sparse attention and a 1 million token context leng…
-
France accelerates AI infrastructure with NVIDIA, Mistral AI builds data center · 1 source tracked
France is significantly advancing its AI infrastructure and ecosystem, with substantial investments and new facilities coming online. Mistral AI is building a large data center in France, leveraging NVIDIA's GB200 syste…
-
New UFP4 recipe tackles shrinkage bias in LLM FP4 pretraining
A new research paper introduces UFP4, a uniform 4-bit training recipe designed to address shrinkage bias in large language model pretraining. The study identifies that current non-uniform FP4 formats, like E2M1 used in …
-
MiniMax M3 integrates with NVIDIA hardware, vLLM, and Inferact
SemiAnalysis reported on the successful integration of MiniMax AI's M3 model with NVIDIA's hardware, specifically highlighting the vLLM project and Inferact's EAGLE3 spec decode. This collaboration focuses on enabling d…
-
DecagonAI cuts voice agent costs 6x with Together AI and open models
DecagonAI has significantly reduced the cost of its voice agent by nearly sixfold by migrating from closed models to fine-tuned open-source models hosted on Together AI. This transition maintained low latency for real-t…
-
NVIDIA Blackwell platform dominates MLPerf Training 6.0 benchmarks · 4 sources tracked
NVIDIA's Blackwell platform has achieved top performance across all seven benchmarks in the MLPerf Training 6.0 industry standard tests. The platform demonstrated the fastest training times and enabled the largest-scale…
-
Tensordyne unveils log-math AI chips, claiming 17x power efficiency
Tensordyne, a startup, has introduced a new AI accelerator that utilizes logarithmic mathematics to improve efficiency. This approach, which rewrites multiplication as addition, claims to offer a 17-fold increase in per…
-
AI Infrastructure Deals Shift Focus to GPU Utilization Over Capacity
QumulusAI has secured over $124 million in three-year agreements focused on AI infrastructure utilization, particularly for inference workloads. This shift indicates a growing customer priority on efficiently using GPU …
-
NVIDIA Blackwell Systems Lead New Agentic AI Benchmarks
NVIDIA has set new performance records on the first agentic AI benchmarks, AgentPerf and Agentic AI Benchmark. The company's GB300 NVL72 system, powered by Blackwell architecture, demonstrated up to a 20x performance le…
-
NVIDIA GeForce NOW offers major savings on cloud gaming memberships
NVIDIA is currently holding a summer sale for its GeForce NOW cloud gaming service, offering significant discounts on 12-month memberships. The sale aims to attract new users by highlighting the convenience of playing P…
-
Apple to use Nvidia chips on Google Cloud for AI Siri
Apple is reportedly planning to integrate Nvidia's Blackwell B200 chips within Google Cloud's infrastructure to power an upcoming AI-enhanced version of Siri. This move is intended to overcome limitations with Apple's i…
-
SANA-Streaming enables real-time video editing on consumer GPUs
Researchers have developed SANA-Streaming, a framework for real-time video editing on consumer GPUs. It utilizes a hybrid diffusion transformer architecture with attention mechanisms for improved local modeling and effi…
-
New Cognitive Kardashev Scale measures civilization's AI computation potential
Researchers have proposed a new framework called the Cognitive Kardashev Scale to measure the computational capacity of civilizations. This scale, analogous to the power-based Kardashev scale, quantifies the amount of s…
-
Developer creates SM1, a memory-efficient Mamba variant for PyTorch
A developer has created SM1, a variant of the Mamba1 architecture, optimized for PyTorch and capable of running on NVIDIA Blackwell hardware. SM1 replaces the selective scan with two native PyTorch operations, achieving…
-
Together AI launches self-service GPU clusters for AI development
Together AI has launched Together Instant Clusters, a new service providing readily available, self-service GPU clusters for AI development and deployment. This offering aims to simplify the complex process of setting u…
-
NVIDIA GTC Taipei: Vera Rubin NVL72 and Jetson Thor lead AI hardware showcase
NVIDIA is showcasing its latest AI innovations at GTC Taipei, including the Vera Rubin NVL72 AI supercomputer, which received multiple Best Choice Awards. This system is designed for large-scale AI inference and trainin…
-
SemiAnalysis dissects NVIDIA Blackwell B200 GPU architecture
SemiAnalysis has analyzed the physical structure of NVIDIA's Blackwell B200 GPU, revealing details about its architecture and manufacturing process. The analysis suggests a complex chiplet design and advanced packaging …