Blackwell GPUs
PulseAugur coverage of Blackwell GPUs — every cluster mentioning Blackwell GPUs across labs, papers, and developer communities, ranked by signal.
2 day(s) with sentiment data
-
Apache TVM launches TIRx compiler for evolving ML kernels and hardware
Apache TVM has launched TIRx, an open-source compiler stack designed for machine learning kernels and evolving hardware. This new system allows for hardware-native DSLs and compilation to GPUs and specialized AI acceler…
-
New speculative decoding methods boost LLM inference speed and safety
Researchers are developing advanced speculative decoding techniques to accelerate large language model inference. HyperDFlash optimizes decoding for DeepSeek-V4's multi-hyper-connection architecture, improving draft acc…
-
Unsloth updates AI library with CUDA 13.3, Mac, and Windows support
Unsloth has released updates for its AI optimization library, including support for CUDA 13.3 on Windows and improved compatibility for Apple Silicon Macs. The new version addresses issues with CUDA 13.2 and provides pr…
-
Together AI releases FlashAttention-3 and -4 for faster LLM processing
Together AI has released FlashAttention-3 and FlashAttention-4, significant upgrades to their GPU-accelerated attention mechanism for large language models. FlashAttention-3, designed for Hopper GPUs, achieves up to 75%…
-
ThriftAttention boosts AI efficiency with selective mixed-precision attention
Researchers have developed ThriftAttention, a novel method to improve the efficiency of long-context attention mechanisms in AI models. This technique selectively applies higher precision (FP16) to a small percentage of…
-
LongLive-2.0 infrastructure accelerates long video generation training and inference
Researchers have developed LongLive-2.0, a parallel infrastructure designed to optimize the training and inference of long video generation models. This system utilizes NVFP4 precision and sequence-parallel autoregressi…
-
Nvidia's RTX Mega Geometry boosts path tracing with VRAM reduction
Nvidia has developed RTX Mega Geometry, a new technology designed to enhance path-traced rendering by reducing VRAM consumption and eliminating visual artifacts. This innovation addresses limitations in existing APIs li…
-
AMD and OpenAI boost 2026 AI performance with new chips and GPUs
AMD has announced new Ryzen AI PRO chips for 2026, designed to boost on-device AI performance and security for enterprise users. Separately, OpenAI has revealed a new training specification utilizing NVIDIA's Blackwell …
-
LongCat Image Edit and SageAttention boost AI image generation speed and realism
A new model called LongCat Image Edit 2026, developed by Meituan, demonstrates superior facial inpainting capabilities in Stable Diffusion workflows, achieving results with 30% greater speed and zero artifacts. This mod…
-
Thinking Machines names Soumith Chintala new CTO, forms vLLM team
Thinking Machines has appointed Soumith Chintala as its new Chief Technology Officer, succeeding Barret Zoph. Chintala, who has a decade of experience in AI research and development, is expected to lead the company's te…