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

Home/Benchmarks

Timing comparison on SocialIQA

60MHA

Mistral

Updated 1mo ago

Evaluation Results

Method	Links
Mistral 2026.04		60	46
GPT 2026.04		33.3	15

SOTA Paper

Mistral

Comparing Human and Large Language Model Interpretation of Implicit Information

Dataset

SocialIQA

Follow for update

@wizwand_team Discord

Related Benchmarks

Inference correction review (discard) on SocialIQA Inference correction review (discard) on COPA Triplet classification on SocialIQA Event/state classification on SocialIQA Inference correction review (reason) on SocialIQA

© 2026 wizwand

Blog Contact Changelog Swarm

Privacy Policy Terms of Service FAQs Swarm Docs