Benchmark family

Swe Bench

A family groups the versions of one benchmark. It is not evaluated and has no applicable models of its own; each of its 4 definitions is resolved separately.

Software Engineering

Definitions

Compare versions →

Newest first, following the recorded successor chain. Each definition is resolved against models separately.

BenchmarkSWE-bench
ActiveNewest in family
Scale
12 repositorys
Primary metric
Resolved Rate
Applicable models
6
BenchmarkSWE-bench Lite
Active
Subset of SWE-bench
Scale
23 tasks
Primary metric
Resolved Rate
Applicable models
6
BenchmarkSWE-bench Multilingual
Active
Derived from SWE-bench
Scale
42 repositorys
Primary metric
Resolved Rate
Applicable models
6
BenchmarkSWE-bench Verified
Active
Subset of SWE-bench
Scale
500 tasks
Primary metric
Resolved Rate
Applicable models
6

Related outside this family

ABC-Bench — related to — SWE-bench

CCBench — related to — SWE-bench

CodeClash — related to — SWE-bench

Multi-SWE-bench — derived from — SWE-bench

Multi-SWE-bench — related to — SWE-bench Multilingual

ProgramBench — related to — SWE-bench

RepoReason — related to — SWE-bench

RuBench 1.0 — related to — SWE-bench

SWE-bench-Live — derived from — SWE-bench

SWE-bench Pro — derived from — SWE-bench

SWE-Bench ProMax v1 — related to — SWE-bench

SWE-bench Science — related to — SWE-bench

SWE-Explore — derived from — SWE-bench Multilingual

SWE-Explore — derived from — SWE-bench Verified

SWE-Marathon v1.1 — related to — SWE-bench

SWE-PolyBench — related to — SWE-bench

SWE-Prometheus v1.0 — related to — SWE-bench

SWE-Proof — derived from — SWE-bench Verified

SWE-Together — related to — SWE-bench

Family methodology

No family-level methodology recorded.

Across definitions

Tasks: Issue Resolution
Sources: None recorded at family level
Last reviewed: 2026-09-23

Applicability does not mean evaluated; benchmark result/model score data is outside this UI contract.