whatmodel.sh / decision / ZTC-Judge-9B (DecisionBench 05d154a5)

ZTC-Judge-9B (DecisionBench 05d154a5)

+ contribute
Update model informationAdd a place to run itAdd a test resultPropose a benchmarkCorrect a benchmark scoreEdit data on GitHub

ZTC-Judge-9B (DecisionBench 05d154a5) is an open-weights decision model from FINAL-Bench. Its published size is 9.653104368B parameters. Licence: apache-2.0.

Catalogue identity notes

DecisionBench v1 comparison: FINAL-Bench/ZTC-Judge-9B, revision 05d154a56b6d48ea7f2d999cabbafae3f26734b9; classifier; adapter jev-compatible-server-v1; probability source native_probability_distribution. Kept separate from newer or unpinned Decision Index recipes; standard prompt, not compact. No unverified runtime or OpenRouter price imported. Current repository metadata checked separately; an unpinned alias is not evidence that today’s weights reproduce this historical run.

Model information sources

· Weights licence ↗ · Parameter count ↗

Published served parameter count; not a RAM or download-size measurement.

51.1±1.6#194 of 355 decision modelsfrom 27 benchmark results across 1 source

$ how it compares

The decision models ranked either side of ZTC-Judge-9B (DecisionBench 05d154a5), out of 357 with a score, then more from FINAL-Bench.

#modelscorewhereRAMreleased
194Aplomb 1 EmpirioLabs AI52±1request access
195Shisa DE-2 (repeat) Shisa.AI51±4model files
196ZTC-Judge-9B (DecisionBench 05d154a5) FINAL-Bench51±2model files
197Jev-Omni akhilaaa351±1model files2026-09-22
198LiquidAI d1-3B Liquid AI51±1model files
more from FINAL-Bench
320ZTC-Judge-4B (DecisionBench b0bb1603) FINAL-Bench29±2model files

see all decision models · find the ones that fit your machine

$ what it's good at

workflows▁▁▁▂▁▂▂▆█▄▁▁ #74 of 114 · 99% of its score · 27 tests

Trails Solar Decide Flash, InternLM Intern-Decision Qwen3.5-4B, StartLux-Decision v1.1 0.8B here, though they rank lower overall.

testresultvs the others in its rundifficultycountswhy
DecisionBench v1 · action selection (all-request accuracy) submitted50.1% ✎▂▁▂▃▄▃▁▁▃█▇▄ beats 27/56medium · avg 51%▇▇▇▇▇▇▇▇ 19%at its level
DecisionBench v1 · routing (all-request accuracy) submitted55.0% ✎▁▁▁▁▂▂▁▁▃▁▃█ beats 22/56medium · avg 64%▇▇▇▇▇▇ 14%sharp · at its level
DecisionBench v1 · navigation (all-request accuracy) submitted68.0% ✎▁▁·▃▁▁▁▁▁▁▃█ beats 31/56medium · avg 58%▇▇▇▇ 11%sharp · at its level
DecisionBench v1 · record classification (all-request accuracy) submitted37.0% ✎▃▂▁▂▂▂▄▁▃▆██ beats 25/56medium · avg 39%▇▇▇ 8.4%at its level
DecisionBench v1 · verification (all-request accuracy) submitted63.0% ✎▁▁▅▁▃▂▂▁▁▇▇█ beats 51/56medium · avg 56%▇▇ 6.1%blunt · at its level
DecisionBench v1 · completion assessment (all-request accuracy) submitted94.0% ✎▁▁▂▃▄·▂▁▁▆█▂ beats 52/56very easy · avg 86%▇▇ 5.0%sharp
DecisionBench v1 · document record classification (all-request accuracy) submitted44.6% ✎▂▂·▁▂▄▃▆▄█▁▁ beats 28/56medium · avg 46%▇▇ 4.8%at its level
DecisionBench v1 · triage (all-request accuracy) submitted81.0% ✎▁·▁·▂▃▄▃▂█▄▂ beats 33/56easy · avg 78%▇▇ 4.8%sharp
19 more tests
testresultvs the others in its rundifficultycountswhy
DecisionBench v1 · bounded extraction (all-request accuracy) submitted93.8% ✎▃▃▂·▁▁▁▁▁▁█▅ beats 37/56very easy · avg 86%▇▇ 4.6%sharp · noisy
DecisionBench v1 · rubric scoring prioritization (all-request accuracy) submitted42.3% ✎▁▁▅▅▃▃▅█▃▃·▁ beats 27/56medium · avg 45%▇▇ 4.1%at its level
DecisionBench v1 · content classification (all-request accuracy) submitted75.0% ✎▁····▁▁▁▁▁▁█ beats 26/56easy · avg 73%▇ 3.5%at its level
DecisionBench v1 · review scoring (all-request accuracy) submitted46.0% ✎▁······▂▃▅▇█ beats 23/56medium · avg 52%▇ 2.7%at its level
DecisionBench v1 · retrieval ranking (all-request accuracy) submitted35.0% ✎▁▁·▁▁▂▃▂▃▃▁█ beats 32/56medium · avg 35%▇ 2.0%at its level · noisy
DecisionBench v1 · risk scoring (all-request accuracy) submitted55.0% ✎▁▁·▁▂▂▁▃▃█▇▃ beats 18/56medium · avg 62%▇ 1.9%blunt · at its level
DecisionBench v1 · target selection (all-request accuracy) submitted39.0% ✎▁▁·▁▁▁▃▁▂▁▃█ beats 34/56hard · avg 29%▇ 1.9%at its level · noisy
DecisionBench v1 · reasoning (all-request accuracy) submitted41.3% ✎▂·▁▁▆█▅▃▃▂▁▁ beats 23/56medium · avg 52%▇ 1.2%at its level
DecisionBench v1 · retrieval verification (all-request accuracy) submitted86.3% ✎█▁▂▁▁▁▁▁▂▅▁▁ beats 40/56easy · avg 74%▇ 0.8%at its level · noisy
DecisionBench v1 · document workflows (all-request accuracy) submitted56.8% ✎█▄▃▁▁▃▂▂▁·▁▁ beats 28/56easy · avg 66%▇ 0.7%blunt · at its level
DecisionBench v1 · guardrails moderation (all-request accuracy) submitted61.5% ✎▁▁▁▂▆▃▃█···▁ beats 37/56medium · avg 52%▇ 0.5%blunt · at its level · noisy
DecisionBench v1 · safety gating (all-request accuracy) submitted97.0% ✎▂·····▆█▂▂▄▇ beats 51/56easy · avg 77%▇ 0.4%noisy
DecisionBench v1 · document classification (all-request accuracy) submitted37.0% ✎▂▁·▂▁▁▁▄▁▂▂█ beats 52/56hard · avg 21%▇ 0.3%blunt · noisy
DecisionBench v1 · semantic filtering (all-request accuracy) submitted71.0% ✎▁·▁▁·▁▁█▂▃█▂ beats 27/56easy · avg 71%▇ 0.3%blunt · at its level · noisy
DecisionBench v1 · content moderation (all-request accuracy) submitted95.0% ✎▁····▃▅▃▂▃·█ beats 35/56easy · avg 82%▇ 0.2%noisy
DecisionBench v1 · context management (all-request accuracy) submitted87.0% ✎▁·····▁▇▆▁█▃ beats 37/56easy · avg 73%▇ 0.2%blunt · at its level · noisy
DecisionBench v1 · entity alignment (all-request accuracy) submitted69.2% ✎▁·▁▄▇▄▅▄▂▁▄█ beats 27/56easy · avg 66%▇ 0.2%at its level · noisy
DecisionBench v1 · function agent skill routing (all-request accuracy) submitted74.8% ✎▃▃▄▅▆▂▅▄▃▄█▃ beats 28/56easy · avg 68%▇ 0.1%blunt · at its level · noisy
DecisionBench v1 · routing triage (all-request accuracy) submitted52.0% ✎█▁···▁·▂▆▅▂▂ beats 31/56medium · avg 42%▇ 0.0%blunt · at its level · noisy

$ where to run it

Download model files ↗
Published model files verified. This is the revision recorded by the benchmark. A local run command and hardware requirements have not been verified for this entry. Source · checked 2026-10-31.

Model repository and documentation: huggingface.co/FINAL-Bench ↗

$ how the score was worked out · Hanno Labs DecisionBench v1 r… 99%

Each result implies a score on its own, after adjusting for how hard its test is and who ran it. The score is their weighted average, with a small pull toward the middle for thinly tested models. Each test's weight is its counts column above; here it's added up by who ran the tests. How it works

sourceresultsshare of the scorelicence
Hanno Labs DecisionBench v1 reviewed results · submitted; independent execution unverified27CC0 1.0
prior · pull toward the middle
weighted average= 51.1

$ improve this page

Update model informationAdd a place to run itAdd a test resultPropose a benchmarkCorrect a benchmark scoreEdit data on GitHub

Use a short GitHub form or edit the data through a pull request. Every number needs a source; here's which ones we can use.

data updated · sources and licences · json