VulcanBench

VulcanBench is a coding benchmark that evaluates AI models on accuracy and cost efficiency, developed by an unspecified organization. It gained prominence when Grok 4.5 led its leaderboard with a score of 82.7, matching Anthropic's Claude Opus-4.8.

Two-hop subgraph: this entity, every entity it directly relates to, and every entity those neighbors relate to. Drag a node, scroll to zoom, click to inspect — or click any neighbor and re-center the atlas there.