Stubbi
Rankings
Benchmark

AlpacaEval

Estimate instruction-following quality with an automatic judge.

Checked Oct 22, 2026Stanford Tatsu Lab

Explore the original evidence

Explore the original evaluation, its methods, and reported findings.

Open eval

Coverage

805 instructions; AlpacaEval 2.0 compares each answer with a GPT-4 Turbo reference answer.

Methodology

A model judge picks the better answer; report win rate and length-controlled win rate.

People & organizations

5

Sources

Length-controlled AlpacaEval paper