SWE-bench Multilingual

The 'SWE-bench Multilingual' is a benchmark dataset used to evaluate the performance of coding agents on software engineering tasks across multiple languages.

Papers using SWE-bench Multilingual (9)

Laguna M.1/XS.2 Technical Report2026From SWE-ZERO to SWE-HERO: Execution-free to Execution-based Fine-tuning for Software Engineering Agents2026Composer 2 Technical Report2026SWE Context Bench: A Benchmark for Context Learning in Coding2026SWE-Bench++: A Framework for the Scalable Generation of Software Engineering Benchmarks from Open-Source Repositories2025SWE-Bench++: A Framework for the Scalable Generation of Software Engineering Benchmarks from Open-Source Repositories2025Composer 2 Technical Report2026Claw-SWE-Bench: A Benchmark for Evaluating OpenClaw-style Agent Harnesses on Coding Tasks2026FastContext: Training Efficient Repository Explorer for Coding Agents2026