AI
Open-Weight Models Are Closing the Gap With Frontier Labs, New Benchmarks Show
A fresh round of independent benchmarking shows the best open-weight language models within striking distance of frontier closed models on reasoning tasks, narrowing a gap that used to be measured in years.
Devon Achterberg//2 min