ChainSWE Benchmark

What is ChainSWE?

ChainSWE evaluates coding agents on sequential, dependent bug fixes within a shared codebase. It includes 304 issues across 54 Python projects, forming…

Released

2026-07-01

Evaluates

Code & Software, Software & AI Compute, cs.SE

Openness

unknown

Importer

Claire Radar

Review status

unreviewed

Reported scores

0

Source provenance

• Original evidence https://arxiv.org/abs/2607.02606

ChainSWE paper

• Paper https://arxiv.org/abs/2607.02606

Which sources cite ChainSWE?

• ChainSWE

Related Code & Software, Software & AI Compute, cs.SE benchmarks

• Active-SWE
• AFT-Bench
• Agent App Benchmark
• Agent Retrieval Bench
• Agent Systems Benchmark (ASB)
• Agentic Evaluation in 2D Mazes
• Agentic Security Benchmark
• AgentInfraBench