0

EvalPlus Team

Academic research team behind the EvalPlus framework - rigorous test-augmented evaluation of code generation (HumanEval+ / MBPP+).

Visit
Type
research group
HQ
USA (UIUC + collaborators)
Founded
2023
Funding
Academic; led by Jiawei Liu (UIUC) and collaborators
Website
evalplus.github.io/leaderboard.html
GitHub
github.com/evalplus/evalplus

Cite

Notes

Only stored in your browser.

Evals
2
Tools
0
Models
0
Papers
0
Boards
1
People
0

Leaderboards

1
EvalPlus LeaderboardThe EvalPlus team's joint HumanEval+ / MBPP+ leaderboard ranking LLMs on code-generation benchmarks expanded with 35-80× more test cases.

Evals

2

HumanEval+

Extended HumanEval with 80× more test cases (EvalPlus). Catches more edge-case bugs than the original.

MBPP+

Extended MBPP with 35× more test cases (EvalPlus). Harder than vanilla MBPP.