Share your thoughts, 1 month free Claude Pro on usSee more

Home/Benchmarks

Pairwise Preference Prediction on SWT-bench

90.11Accuracy

PACE

Updated 22d ago

Evaluation Results

Method	Links
PACE 2026.07		90.11

SOTA Paper

PACE

PACE: A Proxy for Agentic Capability Evaluation

Dataset

SWT-bench

Follow for update

@wizwand_team Discord

Related Benchmarks

Absolute Score Prediction on GAIA Pairwise Preference Prediction on GAIA Absolute Score Prediction on SWE-bench Verified Pairwise Preference Prediction on SWE-bench Verified Absolute Score Prediction on SWE-bench Multimodal

© 2026 wizwand

Blog Contact Changelog Swarm

Privacy Policy Terms of Service FAQs Swarm Docs