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

Home/Benchmarks

Puzzle Solving on KK

50.14Pass@1 Accuracy

Block-R1

Updated 21d ago

Evaluation Results

Method	Links
Block-R1 2026.05		50.14
StableDRL 2026.05		37.57

SOTA Paper

Block-R1

Block-R1: Rethinking the Role of Block Size in Multi-domain Reinforcement Learning for Diffusion Large Language Models

Dataset

Kk

Follow for update

@wizwand_team Discord

Related Benchmarks

Logical Reasoning on KK Mathematical Reasoning on MATH500 (Test accuracy (%))Code Generation on MBPP (Test accuracy %)Puzzle Solving on Sudoku (Test accuracy (%))Code Generation on HumanEval (pass@1 accuracy)

© 2026 wizwand

Blog Contact Changelog Swarm

Privacy Policy Terms of Service FAQs Swarm Docs