Benchmark

RepoQA

RepoQA is a benchmark for evaluating long-context code understanding capabilities of Large Language Models through the Searching Needle Function (SNF)…

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

Reported scores

llm_stats

ModelOrganizationReported valueReportedEvidence
Phi-3.5-mini-instructMicrosoft0.772024-08-23evidence
Phi-3.5-MoE-instructMicrosoft0.852024-08-23evidence

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.