Benchmark

SWE Atlas - Codebase QnA

SWE Atlas - Codebase QnA evaluates a model's ability to answer questions about real codebases, measuring repository-level comprehension and the ability to…

Modality
text
Categories
agents, code
Openness
unknown
Source
llm_stats
Reported scores
2

Reported scores

llm_stats

ModelOrganizationReported valueReportedEvidence
Laguna S 2.1Poolside0.4622026-07-21evidence
MiniMax M3MiniMax0.3792026-06-01evidence

Scores are partitioned by the source that reported them and are never merged into a single cross-source ranking, because the sources measure different things and say so.