Atlas

Benchmarks

← All benchmarks

FrontierSWE: libexpat to x86-64 Assembly (Mean@5)

Code · 2026-04-16

Reimplement the libexpat XML parser as a drop-in x86-64 assembly shared library. FrontierSWE Mean@5 is the average official task score across up to five independent long-horizon agent runs.

Top models (higher is better)

ModelScore
Opus 4.60.1
GPT-5.50.1
Opus 4.70.1
Opus 4.80.0
Claude Fable 50.0
GLM-5.20.0
Grok 4.50.0
Kimi K2.50.0
Qwen3.6 Plus (2026-04-02)0.0
Composer 2.50.0
DeepSeek-V4-Pro0.0
Gemini 3.1 Pro Preview0.0
GLM-5.10.0
GPT-5.40.0
Kimi K2.60.0
Loading Atlas data…