Artificial Analysis's V5 index is said to use more difficult benchmarks
One poster suggests OpenAI has the best reinforcement learning—or that its models perform poorly without reasoning.
TLDR
A post says Artificial Analysis is using more difficult benchmarks in its V5 index. The same author offers two competing takes on OpenAI: it has the best reinforcement learning, or its models are weak without reasoning. Separately, another poster calls a quoted item an “interesting representation of the frontier advantage.”
Combined views
2.2K
4 Sources, first seen ago



