Skip to main content

MapikaSupersonic LabsDecision matchup

Decider 2Bvs.Julia 1

Which model makes better decisions?

Decider 2B does better in independent tests, while Julia 1 is much smaller.

Independent tests
Decider 2Bahead2 of 2
Julia 10 of 2
Jev Decision Index, DecideBench
Popularity
Decider 2Bahead52
Julia 139
out of 100
Weights and access
Decider 2Bopen, local
Julia 1open, local
Size
Decider 2B1.9 B
Julia 1smaller0.14 B
Parameters

Pick Decider 2B for

Jev Decision IndexDecideBenchmore popular

Pick Julia 1 for

smaller

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”.

Jev Decision Index

Decider 2B ahead

Decider 2B is clearly ahead in 19 of 37 subtests and slightly in 12, Julia 1 in 0 clearly and slightly in 1. 5 are level.

SATA-Bench · Decision Index 0.4.1

Decider 2B24.7
Julia 11.0

1,650 tasks, as of October 28, 2026

HellaSwag · Decision Index 0.4.1

Decider 2B85.8
Julia 129.9

10,042 tasks, as of October 28, 2026

MMLU-Pro · Decision Index 0.4.1

Decider 2B37.6
Julia 114.7

12,032 tasks, as of October 28, 2026

Show all 37 subtests
SubtestDecider 2BJulia 1Verdict
SATA-Bench · Decision Index 0.4.124.71.0Decider 2B ahead
HellaSwag · Decision Index 0.4.185.829.9Decider 2B ahead
MMLU-Pro · Decision Index 0.4.137.614.7Decider 2B ahead
When2Call · Decision Index 0.4.154.723.6Decider 2B ahead
BFCL · Decision Index 0.4.182.937.5Decider 2B ahead
New Yorker · Decision Index 0.4.144.723.5Decider 2B ahead
API-Bank · Decision Index 0.4.131.718.3Decider 2B ahead
BBH · Decision Index 0.4.151.033.8Decider 2B ahead
GPQA Diamond · Decision Index 0.4.136.225.5Decider 2B ahead
MuSR · Decision Index 0.4.154.939.0Decider 2B ahead
GSM8K · Decision Index 0.4.123.116.9Decider 2B ahead
WinoGrande · Decision Index 0.4.167.750.7Decider 2B ahead
BPoMP · Decision Index 0.4.164.948.8Decider 2B ahead
ChessBench · Decision Index 0.4.19.97.5Decider 2B ahead
Humicroedit · Decision Index 0.4.159.948.3Decider 2B ahead
Habermas · Decision Index 0.4.142.235.4Decider 2B ahead
HoVer · Decision Index 0.4.159.150.3Decider 2B ahead
CLadder · Decision Index 0.4.157.350.2Decider 2B ahead
cfcolor · Decision Index 0.4.157.050.4Decider 2B ahead
CLINC150 · Decision Index 0.4.173.712.3Decider 2B slightly ahead
ToolRet · Decision Index 0.4.163.113.6Decider 2B slightly ahead
BRIGHT · Decision Index 0.4.137.19.1Decider 2B slightly ahead
iSarcasmEval · Decision Index 0.4.122.46.5Decider 2B slightly ahead
ACOS · Decision Index 0.4.17.93.1Decider 2B slightly ahead
FinEntity · Decision Index 0.4.176.132.8Decider 2B slightly ahead
ContractNLI · Decision Index 0.4.163.431.8Decider 2B slightly ahead
RAGTruth · Decision Index 0.4.128.352.0Julia 1 slightly ahead
ANLI · Decision Index 0.4.149.428.6Decider 2B slightly ahead
BANKING77 · Decision Index 0.4.172.842.6Decider 2B slightly ahead
Amazon ESCI · Decision Index 0.4.134.720.6Decider 2B slightly ahead
VAST · Decision Index 0.4.151.032.9Decider 2B slightly ahead
NLI4CT · Decision Index 0.4.165.549.7Decider 2B slightly ahead
Home appliances · Decision Index 0.4.13.40.0level
POP909 · Decision Index 0.4.10.30.2level
HLE · Decision Index 0.4.112.014.4level
CRUXEval · Decision Index 0.4.133.038.1level
PhishNChips · Decision Index 0.4.150.149.9level

Source · latest October 28, 2026

DecideBench

Decider 2B ahead

Decider 2B is clearly ahead in 7 of 10 subtests, Julia 1 in none. 3 are level.

DecideBench pair accuracy

Decider 2B41.0 %
Julia 18.0 %

200 tasks, as of October 18, 2026

DecideBench review_sentiment accuracy

Decider 2B70.0 %
Julia 122.0 %

50 tasks, as of October 18, 2026

DecideBench ticket_triage accuracy

Decider 2B64.0 %
Julia 134.0 %

50 tasks, as of October 18, 2026

Show all 10 subtests
SubtestDecider 2BJulia 1Verdict
DecideBench pair accuracy41.0 %8.0 %Decider 2B ahead
DecideBench review_sentiment accuracy70.0 %22.0 %Decider 2B ahead
DecideBench ticket_triage accuracy64.0 %34.0 %Decider 2B ahead
DecideBench overall accuracy63.2 %34.8 %Decider 2B ahead
DecideBench support_intent accuracy96.0 %54.0 %Decider 2B ahead
DecideBench content_moderation accuracy66.0 %38.0 %Decider 2B ahead
DecideBench agent_routing accuracy66.0 %40.0 %Decider 2B ahead
DecideBench claim_support accuracy42.0 %24.0 %level
DecideBench returns_policy accuracy46.0 %28.0 %level
DecideBench action_review accuracy56.0 %38.0 %level

Source · latest October 18, 2026

All data side by side

Decider 2BJulia 1
KindStandaloneClassifier
DeveloperMapikaSupersonic Labs
StatusOpen sourceOpen source
Weightsopenopen
Runs locallyyesyes
Parameters1.9 B0.14 B
LicenseApache-2.0Apache-2.0
Popularity (out of 100)5239