Domains: Coding
Skills: AgenticCodingTool UseReasoning

Four-Factor Integrity Breakdown & Primary-Source Evidence

Originality

Weight 25%
0/10
Reuse Profile: High Reuse (>75%)
"30 open-source repositories spanning Python, Java, Kotlin, Rust, and C++. Each is migrated into its four non-source target languages (120 scored repository-language migrations in all), and every migration is graded against a hidden suite of roughly 100 to 400 behavior tests, exercised through the program’s command-line interface."

Freshness

Weight 20%
10/10
Initial Release: 2026-09Latest Refresh: 2026-11Rolling Updates: No
"v1.1 runs every model on the Pi agent harness instead of mini-SWE-agent, and grades against revised hidden test suites. The tasks are the same, but scores are not directly comparable to"

Grading

Weight 15%
10/10
Integrity: Independent Offline (Leak-Verified)Sandbox: offline isolated
"The model works in an offline sandbox with the source program and task instructions, and submits a target-language implementation plus a Dockerfile. The grader builds that image and runs a hidden behavior-test suite the model never sees. Before scoring, a separate anti-cheat checker inspects each submission and zeroes any that wrap the original executable, copy reference artifacts, or hardcode expected outputs."

Frontier Model Standings on Code Migration v1.1

Top 15 active models tracked
Rank Model Official Score Percentile
#1Claude Sonnet 5.5 max Anthropic86.9%96.6p
#2Claude Fable 5.2 max Anthropic83.3%92.6p
#3Claude Opus 5.5 max Anthropic81.0%88.5p
#4GPT-6 Astra max OpenAI80.6%84.5p
#5GPT-6.1 Sol max OpenAI77.3%80.4p
#6Muse Wonder max Meta72.4%76.4p
#7Gemini 4 Argon high Google70.8%72.4p
#8DeepSeek V4.1 Flash 1017 max DeepSeek67.5%68.4p
#9Muse Wonder Full max Meta62.8%64.3p
#10Qwen3.8 Max (0902) xhigh Alibaba57.2%60.3p
#11GLM-5.3 max Z AI55.1%56.2p
#12Grok 4.7 xhigh SpaceXAI53.2%52.2p
#13Claude Haiku 5.5 max Anthropic52.4%48.1p
#16MiMo-V2.6-Pro default Xiaomi48.8%36.0p
#17Step 5 Preview default StepFun47.0%31.9p

Embeddable Audit Badge

View Markdown (.md)

Benchmark maintainers can embed this live SVG badge in GitHub READMEs or documentation:

BenchAudit Code Migration v1.1 Tier A 7.5/10
[![BenchAudit Tier A](https://benchaudit.org/badges/code-migration-v1-1.svg)](https://benchaudit.org/benchmarks/code-migration-v1-1/)