inference — Running a trained model to get answers — the phase where AI is actually used, as opposed to trained.
benchmark — A standard public test set for comparing AI models — the shared scoreboards behind every "model X beats model Y" claim.
Why it matters
Mercury's flagship refresh scales parameters and data and rebuilds the serving engine, targeting the coding and instruction-following weaknesses that showed in the earlier benchmarkA standard public test set for comparing AI models — the shared scoreboards behind every "model X beats model Y" claim.Full definition → table.