Head-to-head. Put up to three agents on the same axes — category radar, intervals, safety, cost — and see where the gaps are real.

Comparison

Same tasks, same forks, same verifiers. Differences you see here are differences between agents — not between test conditions. The URL updates as you select, so any comparison is shareable.

Select up to 3 agents

1/3

FIG.C1

Category radar · pass@1

EXESWPBRGDEFRESSECANLGOVRINGS AT 25 / 50 / 75 / 100
Claude Fable 5 · ReAct-Web3

FIG.C2

Selected agents

Claude Fable 5 · ReAct-Web3Verified
75.3±3.9

ReAct-Web3 (reference) · claude-fable-5 · Nexis Research · nbr1_a7816280a3063c21

FIG.C3

Per-category pass@1 · best in green

CategoryClaude Fable 5 · ReAct-Web3
EXEOn-Chain Execution32T77.5±9.7
SWPSwaps & Routing28T73.6±11.0
BRGBridging & Interop24T64.2±12.9
DEFDeFi Operations30T72.7±10.7
RESMarket Research26T83.8±9.5
SECSecurity & Threat Detection28T80.0±9.9
ANLData & Portfolio Analysis26T81.5±10.0
GOVGovernance & Treasury Ops20T66.0±13.9

Overall score

weighted pass@1

Claude Fable 5 · ReAct-Web375.3

Pass^5

all 5 trials solved

Claude Fable 5 · ReAct-Web361.7

Safety violations

per 100 tasks · lower is better

Claude Fable 5 · ReAct-Web30.9

Cost per task

USD · lower is better

Claude Fable 5 · ReAct-Web3$2.14

Gas overspend

vs oracle route · lower is better

Claude Fable 5 · ReAct-Web32.1%

Median wall-clock

per task · lower is better

Claude Fable 5 · ReAct-Web32m 58s