II.
Benchmark overview
Reference · livebenchmark:mbpp
MBPP overview
Mostly Basic Python Problems — ~1,000 short Python programming problems with tests.
Attributes
displayName
MBPP
homepageUrl
kind
code-generation
targetsKind
ModelVersion
description
Mostly Basic Python Problems — ~1,000 short Python programming problems with tests.
Outgoing edges
None.
Incoming edges
belongs_to_benchmark1
- test-set:mbpp-full·TestSetMBPP full problem set
bounds_subject1
- scope-boundary:mbpp.scope·ScopeBoundary
for_benchmark1
scored_against1
- eval-result:mbpp.qwen-2-5-coder-32b.001·EvalResult