whatmodel.sh / decision / OpenDecision (DecisionBench a51e07b5)

OpenDecision (DecisionBench a51e07b5)

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

OpenDecision (DecisionBench a51e07b5) is an open-weights decision model from deepanwadhwa. Its published size is 0.395833346B parameters. Licence: apache-2.0.

Catalogue identity notes

DecisionBench v1 comparison: deepanwadhwa/OpenDecision, revision a51e07b524299e309dd2b88d48b0cfa2bd9ec598; classifier; adapter jev-compatible-server-nli-entailment-v1; probability source softmax-normalized-entailment-logits-over-candidates-v1. 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.

29.8±1.6#309 of 355 decision modelsfrom 27 benchmark results across 1 source

$ how it compares

The decision models ranked either side of OpenDecision (DecisionBench a51e07b5), out of 357 with a score.

#modelscorewhereRAMreleased
309PICK-150M v5 lCREEl30±1model files2026-10-18
310InternLM Intern-Decision Qwen3.5-0.8B Shanghai AI Laboratory (InternLM)30±1model files
311OpenDecision (DecisionBench a51e07b5) deepanwadhwa30±2model files
312Crivo-68M tardellirs30±1model files
313GLiNER2.5-Decide (DecisionBench 0872ab14) Fastino30±2model files

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

$ what it's good at

workflows▁▁▁▃▁▂▂▆█▄▁▁ #94 of 114 · 99% of its score · 27 tests

Trails GLiNER2.5-Decide, vLLM-SR Decision 2.0 Kai 0.6B, LiquidAI d1-omni-600M here, though they rank lower overall.

testresultvs the others in its rundifficultycountswhy
DecisionBench v1 · action selection (all-request accuracy) submitted28.4% ✎▂▁▂▃▄▃▁▁▂█▇▄ beats 15/56medium · avg 51%▇▇▇▇▇▇ 15%at its level
DecisionBench v1 · routing (all-request accuracy) submitted32.3% ✎▁▁▁▁▂▃▁▁▁▁▃█ beats 14/56medium · avg 64%▇▇▇▇▇ 13%sharp · at its level
DecisionBench v1 · completion assessment (all-request accuracy) submitted51.0% ✎▁▁▂▃▄·▂▁▁▆█▂ beats 12/56very easy · avg 86%▇▇▇▇ 10%sharp · at its level
DecisionBench v1 · bounded extraction (all-request accuracy) submitted75.0% ✎▃▃▂·▁▁▃▁▁▁█▅ beats 21/56very easy · avg 86%▇▇▇▇ 10%sharp · at its level · noisy
DecisionBench v1 · navigation (all-request accuracy) submitted15.0% ✎▁▁·▃▁▁▁▁▁▁▃█ beats 5/56medium · avg 58%▇▇▇ 8.7%sharp · at its level
DecisionBench v1 · triage (all-request accuracy) submitted58.0% ✎▁·▁·▂▃▄▃▂█▄▂ beats 17/56easy · avg 78%▇▇▇ 6.9%sharp · at its level
DecisionBench v1 · record classification (all-request accuracy) submitted21.0% ✎▃▂▁▂▃▂▄▁▃▆██ beats 11/56medium · avg 39%▇▇ 6.1%typical
DecisionBench v1 · verification (all-request accuracy) submitted41.5% ✎▁▁▅▁▃▂▂▁▁▇▇█ beats 13/56medium · avg 56%▇▇ 5.8%blunt · at its level
19 more tests
testresultvs the others in its rundifficultycountswhy
DecisionBench v1 · content classification (all-request accuracy) submitted46.0% ✎▁····▁▁▁▁▃▁█ beats 11/56easy · avg 73%▇▇ 4.3%at its level
DecisionBench v1 · document record classification (all-request accuracy) submitted28.7% ✎▂▂·▁▃▄▃▆▄█▁▁ beats 9/56medium · avg 46%▇▇ 3.9%at its level
DecisionBench v1 · rubric scoring prioritization (all-request accuracy) submitted26.1% ✎▁▁▅▅▃▃▅█▃▃·▁ beats 10/56medium · avg 45%▇ 3.3%at its level
DecisionBench v1 · review scoring (all-request accuracy) submitted35.0% ✎▁······▂▃▅▇█ beats 12/56medium · avg 52%▇ 2.4%at its level
DecisionBench v1 · risk scoring (all-request accuracy) submitted57.0% ✎▁▁·▁▂▂▁▃▃█▇▃ beats 22/56medium · avg 62%▇ 1.9%blunt · at its level
DecisionBench v1 · retrieval ranking (all-request accuracy) submitted17.0% ✎▁▁·▁▁▂▃▃▃▁▁█ beats 22/56medium · avg 35%▇ 1.0%noisy
DecisionBench v1 · reasoning (all-request accuracy) submitted34.8% ✎▂·▁▁▆█▅▃▃▂▁▁ beats 12/56medium · avg 52%▇ 1.0%at its level
DecisionBench v1 · target selection (all-request accuracy) submitted9.0% ✎▁▁·▁▁▁▃▁▂▁▁█ beats 15/56hard · avg 29%▇ 1.0%noisy
DecisionBench v1 · retrieval verification (all-request accuracy) submitted61.4% ✎█▁▃▁▁▁▁▁▂▅▁▁ beats 27/56easy · avg 74%▇ 0.9%at its level · noisy
DecisionBench v1 · document workflows (all-request accuracy) submitted57.7% ✎█▄▃▁▁▃▂▂▁·▁▁ beats 30/56easy · avg 66%▇ 0.8%blunt · at its level
DecisionBench v1 · safety gating (all-request accuracy) submitted60.0% ✎▂·····▆█▂▂▄▇ beats 22/56easy · avg 77%▇ 0.6%at its level · noisy
DecisionBench v1 · guardrails moderation (all-request accuracy) submitted31.5% ✎▁▁▁▃▆▃▃█···▁ beats 6/56medium · avg 52%▇ 0.5%blunt · at its level · noisy
DecisionBench v1 · semantic filtering (all-request accuracy) submitted83.0% ✎▁·▁▁·▁▁█▂▃█▂ beats 41/56easy · avg 71%▇ 0.3%blunt · at its level · noisy
DecisionBench v1 · content moderation (all-request accuracy) submitted70.0% ✎▁····▃▅▃▂▂·█ beats 21/56easy · avg 82%▇ 0.3%at its level · noisy
DecisionBench v1 · document classification (all-request accuracy) submitted27.0% ✎▂▁·▂▁▁▁▄▁▂▃█ beats 35/56hard · avg 21%▇ 0.2%blunt · noisy
DecisionBench v1 · context management (all-request accuracy) submitted91.0% ✎▁·····▁▇▆▁█▃ beats 47/56easy · avg 73%▇ 0.2%blunt · at its level · noisy
DecisionBench v1 · entity alignment (all-request accuracy) submitted83.5% ✎▁·▁▄▇▄▅▄▂▁▄█ beats 34/56easy · avg 66%▇ 0.2%at its level · noisy
DecisionBench v1 · function agent skill routing (all-request accuracy) submitted75.6% ✎▃▃▄▅▆▂▅▄▃▄█▃ beats 29/56easy · avg 68%▇ 0.1%blunt · at its level · noisy
DecisionBench v1 · routing triage (all-request accuracy) submitted33.8% ✎█▁···▁·▃▆▅▂▂ beats 21/56medium · 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/MoritzLaurer ↗

$ 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= 29.8

$ 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