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

2/3

FIG.C1

Category radar · pass@1

EXESWPBRGDEFRESSECANLGOVRINGS AT 25 / 50 / 75 / 100
Claude Fable 5 · ReAct-Web3Nex-T1 Sentinel

FIG.C2

Selected agents

Claude Fable 5 · ReAct-Web3Verified
75.3±3.9

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

Nex-T1 SentinelVerified
76.3±3.8

Nexis Autonomy Stack · Nex-T1 · Nexis Research · nbr1_80c1c990681d4367

Head-to-head

Nex-T1 Sentinel leads 5 of 8 categories, Δ overall -0.9 pts. The 95% intervals overlap — treat this as a statistical tie.

Score / $ · Claude Fable 5: 35.2 · Nex-T1 Sentinel: 88.7

FIG.C3

Per-category pass@1 · best in green

CategoryClaude Fable 5 · ReAct-Web3Nex-T1 Sentinel
EXEOn-Chain Execution32T77.5±9.782.5±8.8
SWPSwaps & Routing28T73.6±11.080.7±9.8
BRGBridging & Interop24T64.2±12.969.2±12.4
DEFDeFi Operations30T72.7±10.777.3±10.1
RESMarket Research26T83.8±9.572.3±11.5
SECSecurity & Threat Detection28T80.0±9.984.3±9.0
ANLData & Portfolio Analysis26T81.5±10.074.6±11.2
GOVGovernance & Treasury Ops20T66.0±13.963.0±14.2

Overall score

weighted pass@1

Claude Fable 5 · ReAct-Web375.3
Nex-T1 Sentinel76.3

Pass^5

all 5 trials solved

Claude Fable 5 · ReAct-Web361.7
Nex-T1 Sentinel64.0

Safety violations

per 100 tasks · lower is better

Claude Fable 5 · ReAct-Web30.9
Nex-T1 Sentinel0.5

Cost per task

USD · lower is better

Claude Fable 5 · ReAct-Web3$2.14
Nex-T1 Sentinel$0.86

Gas overspend

vs oracle route · lower is better

Claude Fable 5 · ReAct-Web32.1%
Nex-T1 Sentinel1.6%

Median wall-clock

per task · lower is better

Claude Fable 5 · ReAct-Web32m 58s
Nex-T1 Sentinel1m 52s