List your company

Scientific Curation Benchmark

81GRAS ScoreBeaver
42.5252.5162.572.49Jun 19, 2026

Evaluation results

MethodLinks
BeaverModel=gpt-5.42026.0681
CodexModel=gpt-5.42026.0658.6
Claude CodeModel=opus-4.72026.0658.5
Claude CodeModel=opus-4.62026.0658
Copilot CLIModel=gpt-5.42026.0657.4
Meta-HarnessModel=gpt-5.42026.0648
Claude CodeModel=haiku-4.52026.0644