whatmodel.sh / decision / Qwen3-Reranker-4B (DecisionBench 22e68366)

Qwen3-Reranker-4B (DecisionBench 22e68366)

+ contribute

Qwen3-Reranker-4B (DecisionBench 22e68366) is an open-weights decision model from Alibaba (Qwen). Its published size is 4.021784576B parameters. Licence: apache-2.0.

Catalogue identity notes

DecisionBench v1 comparison: Qwen/Qwen3-Reranker-4B, revision 22e683669bc0f0bd69640a1354a6d0aebcfeede5; 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.

41.3±1.6#258 of 355 decision modelsfrom 27 benchmark results across 1 source

$ how it compares

The decision models ranked either side of Qwen3-Reranker-4B (DecisionBench 22e68366), out of 357 with a score.

#	model	score	where
258	Cam1 cbuk10001	42±1	model files
259	Gevva E4B davidburhans	42±1	model files
260	Qwen3-Reranker-4B (DecisionBench 22e68366) Alibaba (Qwen)	41±2	model files
261	mini-jev Qwen3-4B (decision readout) r-ms	41±1	source code
262	open-jev (pngwn) pngwn	41±1	model files

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

$ what it's good at

workflows▁▁▁▂▁▃▂▆█▄▁▁ #87 of 114 · 99% of its score · 27 tests

Trails vLLM-SR Decision 2.0 Kai 0.6B, LiquidAI d1-omni-600M, InternLM Intern-Decision Qwen3.5-0.8B here, though they rank lower overall.

test	result	vs the others in its run	difficulty	counts	why
DecisionBench v1 · action selection (all-request accuracy) submitted	45.6% ✎	▂▁▂▃▄▃▁▃▂█▇▄ beats 24/56	medium · avg 51%	▇▇▇▇▇▇▇ 17%	at its level
DecisionBench v1 · routing (all-request accuracy) submitted	50.3% ✎	▁▁▁▁▂▂▁▃▁▁▃█ beats 20/56	medium · avg 64%	▇▇▇▇▇▇ 14%	sharp · at its level
DecisionBench v1 · navigation (all-request accuracy) submitted	40.0% ✎	▁▁·▃▁▁▁▃▁▁▃█ beats 20/56	medium · avg 58%	▇▇▇▇ 11%	sharp · at its level
DecisionBench v1 · bounded extraction (all-request accuracy) submitted	33.5% ✎	▃▃▃·▁▁▁▁▁▁█▅ beats 15/56	very easy · avg 86%	▇▇▇ 7.5%	sharp · at its level · noisy
DecisionBench v1 · completion assessment (all-request accuracy) submitted	81.0% ✎	▁▁▂▃▄·▂▃▁▆█▂ beats 23/56	very easy · avg 86%	▇▇▇ 7.4%	sharp
DecisionBench v1 · record classification (all-request accuracy) submitted	30.0% ✎	▃▂▁▂▂▂▄▁▃▆██ beats 21/56	medium · avg 39%	▇▇▇ 7.0%	at its level
DecisionBench v1 · triage (all-request accuracy) submitted	70.0% ✎	▁·▁·▂▃▄▃▂█▄▂ beats 24/56	easy · avg 78%	▇▇ 5.8%	sharp · at its level
DecisionBench v1 · verification (all-request accuracy) submitted	48.0% ✎	▁▁▅▁▃▂▃▁▁▇▇█ beats 22/56	medium · avg 56%	▇▇ 5.7%	blunt · at its level

19 more tests

test	result	vs the others in its run	difficulty	counts	why
DecisionBench v1 · document record classification (all-request accuracy) submitted	45.3% ✎	▂▂·▁▂▄▃▆▄█▁▁ beats 29/56	medium · avg 46%	▇▇ 4.2%	at its level
DecisionBench v1 · content classification (all-request accuracy) submitted	75.0% ✎	▁····▁▁▁▁▁▁█ beats 25/56	easy · avg 73%	▇▇ 3.9%	at its level
DecisionBench v1 · rubric scoring prioritization (all-request accuracy) submitted	30.3% ✎	▁▁▅▅▃▃▅█▃▃·▁ beats 16/56	medium · avg 45%	▇ 3.6%	at its level
DecisionBench v1 · review scoring (all-request accuracy) submitted	50.0% ✎	▁······▂▃▅▇█ beats 28/56	medium · avg 52%	▇ 2.5%	at its level
DecisionBench v1 · risk scoring (all-request accuracy) submitted	71.0% ✎	▁▁·▁▂▂▁▃▃█▇▃ beats 53/56	medium · avg 62%	▇ 1.8%	blunt · at its level
DecisionBench v1 · retrieval ranking (all-request accuracy) submitted	13.0% ✎	▁▁·▁▁▂▃▃▃▁▁█ beats 18/56	medium · avg 35%	▇ 1.5%	noisy
DecisionBench v1 · target selection (all-request accuracy) submitted	24.0% ✎	▁▁·▁▁▁▃▁▂▃▁█ beats 28/56	hard · avg 29%	▇ 1.3%	noisy
DecisionBench v1 · reasoning (all-request accuracy) submitted	43.5% ✎	▂·▁▁▆█▅▃▃▂▁▁ beats 28/56	medium · avg 52%	▇ 1.1%	at its level
DecisionBench v1 · retrieval verification (all-request accuracy) submitted	60.9% ✎	█▁▃▁▁▁▁▁▂▅▁▁ beats 25/56	easy · avg 74%	▇ 0.8%	at its level · noisy
DecisionBench v1 · document workflows (all-request accuracy) submitted	60.1% ✎	█▄▃▁▁▃▂▂▁·▁▁ beats 33/56	easy · avg 66%	▇ 0.7%	blunt · at its level
DecisionBench v1 · safety gating (all-request accuracy) submitted	81.0% ✎	▂·····▆█▃▂▄▇ beats 32/56	easy · avg 77%	▇ 0.5%	at its level · noisy
DecisionBench v1 · guardrails moderation (all-request accuracy) submitted	45.3% ✎	▁▁▁▂▆▃▃█···▁ beats 26/56	medium · avg 52%	▇ 0.5%	blunt · at its level · noisy
DecisionBench v1 · semantic filtering (all-request accuracy) submitted	69.0% ✎	▁·▁▁·▁▁█▃▃█▂ beats 26/56	easy · avg 71%	▇ 0.3%	blunt · at its level · noisy
DecisionBench v1 · document classification (all-request accuracy) submitted	30.0% ✎	▂▁·▂▁▁▁▄▁▂▂█ beats 37/56	hard · avg 21%	▇ 0.2%	blunt · noisy
DecisionBench v1 · content moderation (all-request accuracy) submitted	92.0% ✎	▁····▃▅▃▃▂·█ beats 32/56	easy · avg 82%	▇ 0.2%	noisy
DecisionBench v1 · context management (all-request accuracy) submitted	70.0% ✎	▁·····▁▇▆▁█▃ beats 27/56	easy · avg 73%	▇ 0.2%	blunt · at its level · noisy
DecisionBench v1 · entity alignment (all-request accuracy) submitted	100.0% ✎	▁·▁▄▇▄▅▄▂▁▄█ beats 51/56	easy · avg 66%	▇ 0.2%	at its level · noisy
DecisionBench v1 · function agent skill routing (all-request accuracy) submitted	98.3% ✎	▃▃▄▅▆▂▅▄▃▄█▃ beats 54/56	easy · avg 68%	▇ 0.1%	blunt · at its level · noisy
DecisionBench v1 · routing triage (all-request accuracy) submitted	71.4% ✎	█▁···▁·▂▆▅▂▂ beats 43/56	medium · 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/Qwen ↗

$ 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

source	results	share of the score	licence
Hanno Labs DecisionBench v1 reviewed results · submitted; independent execution unverified	27	▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇ 99%	CC0 1.0
prior · pull toward the middle		▇ 1%
weighted average		= 41.3

$ improve this page

Update model information Add a place to run it Add a test result Propose a benchmark Correct a benchmark score Edit 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 2026-11-03 02:37 UTC · sources and licences · json