#benchmark
Benchmark: 42 AI articles covering benchmark news, analysis, and research
Articles
MoonMath AI Open-Sources HIP Attention Kernel for AMD MI300X:⭐8
MoonMath AI open-sources a high-performance HIP attention kernel for AMD MI300X, beating AITER v3 across all tested shapes and rounding modes under MIT license.
OpenAI Releases LifeSciBench: A 750-Task Benchmark for⭐6
OpenAI’s LifeSciBench is a 750-task benchmark evaluating AI on real-life science research with expert rubrics, where even top models pass only ~1 in 3 tasks.