CuTe
PulseAugur coverage of CuTe — every cluster mentioning CuTe across labs, papers, and developer communities, ranked by signal.
2 day(s) with sentiment data
-
Byte-level LLM conversion maintains performance but slashes throughput
Researchers have developed a method to convert a 27B parameter hybrid-attention LLM, Qwen3.8, into a pure byte-level model. This byte model, trained on self-generated data and encyclopedia text, maintains most of the or…
-
KernelSeek AI agent optimizes CUDA kernels with questions, beats baselines
Researchers have developed KernelSeek, a novel question-driven AI agent designed to optimize CUDA kernels for GPU performance. Unlike existing agents that rely on extensive toolkits or fixed policies, KernelSeek uses ta…
-
New system KernelPro autonomously optimizes GPU kernel code using LLMs
Researchers have developed KernelPro, an autonomous system designed to optimize GPU kernel code for large language models. This system integrates LLM code generation with hardware profiler feedback and specialized analy…
-
LLM framework automates GPU kernel generation, outperforming PyTorch
Researchers have developed CuTeGen, a new framework designed to automate the creation and optimization of high-performance GPU kernels. This agentic system employs a structured workflow of generating, testing, and refin…