Benchmark

Repo Env Benchmark

What is Repo Env?

Repo Env evaluates an agent's ability to set up, configure, and run real repositories, including dependency resolution and environment management.

Modality
text
Evaluates
agents, coding
Openness
unknown
Importer
LLM Stats
Reported scores
2

Source provenance

Repo Env results and reported scores

LLM Stats

ModelOrganizationReported valueReportedEvidence
Seed 2.1 ProByteDance0.552026-06-24evidence
Seed 2.1 TurboByteDance0.4672026-06-24evidence

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.

Which sources cite Repo Env?

Related agents, coding benchmarks