Rockman Benchmark v0.2

What is Rockman Benchmark v0.2?

An evaluation suite of 220 code and software engineering tasks measuring LLM performance via executable tests, with a primary 44-task private split.

Released

2026-09-20

Evaluates

Code, Code & Software, Code generation, Software & AI Compute

Openness

unknown

Importer

Claire Radar

Review status

ai-name-audit-deferred

Reported scores

0

Source provenance

• Original evidence https://github.com/InserLoft/Rockman

Rockman Benchmark v0.2 code

• Code https://github.com/InserLoft/Rockman
• Website https://Inserloft.com/Benchmarks/Rockman

Which sources cite Rockman Benchmark v0.2?

• Rockman Benchmark v0.2

Related Code, Code & Software, Code generation, Software & AI Compute benchmarks

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