TypeSafe AISupersonic LabsDecision matchup
Jevvs.Julia 1
Which model makes better decisions?
Jev does better in independent tests, while Julia 1 is available with open weights.
Independent testsDecideBench
Jevahead1 of 1
Julia 10 of 1
Popularityout of 100
Jev43
Julia 139
Weights and access
Jevn/a
Julia 1openopen, local
SizeParameters
Jevn/a
Julia 10.14 B
Pick Jev for
DecideBench
Pick Julia 1 for
open weightsruns locally
Shared tests
Independent evaluators and leaderboards only, the newest shared snapshot per evaluator. With 95 % ranges a lead counts when they do not overlap; for hit rates, when it holds in a 95 % test. Without any uncertainty figure at most “slightly ahead”.
DecideBench
Jev aheadJev is clearly ahead in 10 of 10 subtests, Julia 1 in none.
DecideBench pair accuracy
Jev96.0 %
Julia 18.0 %
200 tasks, as of October 18, 2026
DecideBench review_sentiment accuracy
Jev98.0 %
Julia 122.0 %
50 tasks, as of October 18, 2026
DecideBench claim_support accuracy
Jev96.0 %
Julia 124.0 %
50 tasks, as of October 18, 2026
Show all 10 subtests
| Subtest | Jev | Julia 1 | Verdict |
|---|---|---|---|
| DecideBench pair accuracy | 96.0 % | 8.0 % | Jev ahead |
| DecideBench review_sentiment accuracy | 98.0 % | 22.0 % | Jev ahead |
| DecideBench claim_support accuracy | 96.0 % | 24.0 % | Jev ahead |
| DecideBench returns_policy accuracy | 96.0 % | 28.0 % | Jev ahead |
| DecideBench ticket_triage accuracy | 98.0 % | 34.0 % | Jev ahead |
| DecideBench overall accuracy | 98.0 % | 34.8 % | Jev ahead |
| DecideBench action_review accuracy | 100.0 % | 38.0 % | Jev ahead |
| DecideBench content_moderation accuracy | 98.0 % | 38.0 % | Jev ahead |
| DecideBench agent_routing accuracy | 98.0 % | 40.0 % | Jev ahead |
| DecideBench support_intent accuracy | 100.0 % | 54.0 % | Jev ahead |
Source · latest October 18, 2026