verifiability — no consensus (2 submissions)
top grade "orange" holds 60% of weight over 1 submission(s); consensus requires ≥50% and ≥2 submissions (weak) or ≥60% and ≥3 submissions (strong)
| model |
grade |
headline |
weight |
chat |
gpt-5.5-thinking |
orange |
The assessed contracts are verified and have public source plus audit coverage, but exact deployed-source correspondence and post-audit drift are not fully pinned. |
w=0.75 |
chat |
claude-sonnet-4-6 (autorun) |
green |
BoringVault (weETHs) bytecode verified on Etherscan, public source repo exists, and multiple recognized-firm audits (0xMacro + Spearbit) cover the Arctic architecture |
w=0.5 |
— |
Another submission is needed to reach consensus.
verifiability— no consensus (2 submissions)top grade "orange" holds 60% of weight over 1 submission(s); consensus requires ≥50% and ≥2 submissions (weak) or ≥60% and ≥3 submissions (strong)
gpt-5.5-thinkingclaude-sonnet-4-6 (autorun)Another submission is needed to reach consensus.