◆ Evals & benchmarks

ByteDance Multi-SWE-bench

Apache-2.0

ByteDance Seed's issue-resolving benchmark across Java, TypeScript, JavaScript, Go, Rust, C and C++ — SWE-bench beyond Python.

source ↗ ★ 362 last push 2025-12-18

Summary

👋 Hi, everyone! We are ByteDance Seed team. You can get to know us better through the following channels👇

from the project README

From the project’s own instructions where it documents any; otherwise a plain clone.

shell
```
$ git clone https://github.com/multi-swe-bench/multi-swe-bench.git
```
The README clones over SSH; this is the same repository over HTTPS.
shell
```
$ cd multi-swe-bench && make install
```
The instances (full, mini and flash) are published on Hugging Face under ByteDance-Seed.

2,638 files 5.1 MB repository

Commands are curated, not yet run by us.

jq — Stream the records one per line to see their shape.

sample.jsonl

$ jq -c . sample.jsonl

DuckDB — Query the file in place with SQL — DuckDB detects CSV, JSON and JSONL by extension.

sample.jsonl

$ duckdb -c 'SELECT * FROM '\''sample.jsonl'\'' LIMIT 10'