whatmodel.sh / decision / mini-jev Qwen3-4B (decision readout)

mini-jev Qwen3-4B (decision readout)

+ contribute

mini-jev Qwen3-4B (decision readout) is a source-available decision implementation from r-ms. Its published size is 4.02B parameters.

Catalogue identity notes

Exact Decision Index tested engine mini-jev; reported model mini-jev Qwen3-4B. Type: inference technique; technique: autoregressive. Published checkpoint/base reference: Qwen/Qwen3-4B-Instruct-2507. Implementation: https://github.com/r-ms/mini-jev. Comparison entry; no current run command, API price or speed inferred from the benchmark.

Model information sources

· Parameter count ↗

Publisher parameter count; includes the served backbone, not just adapter size.

41.2±1.1#259 of 355 decision modelsfrom 38 benchmark results across 1 source

$ how it compares

The decision models ranked either side of mini-jev Qwen3-4B (decision readout), out of 357 with a score.

#	model	score	where	RAM
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
263	Decision-1.0-Eos-0.8B (DecisionBench 3c2d6326) llm-semantic-router	41±1	local	~5.8GB

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

$ what it's good at

tools & automation▁▃▁▁▁▃▃▄▆██▆ #207 of 250 · 6.7% of its score · 5 tests

Beats Cam1 here, though they rank higher overall. Trails JevK5 Lite, Mola-68M, Crivo-68M here, though they rank lower overall.

test	result	vs the others in its run	difficulty	counts	why
Decision Index 0.4.1: BFCL (case exact accuracy) submitted	77.4% ✎	▁▂▁▁▁▃▂▄▅▆█▁ beats 42/246	very easy · avg 92%	▇▇ 4.2%	sharp · noisy
Decision Index 0.4.1: When2Call (accuracy) submitted	50.3% ✎	▁▁▁▂▃▅▅▆█▇█▃ beats 42/246	easy · avg 72%	▇ 1.2%	at its level · noisy
Decision Index 0.4.1: ToolRet (nDCG@10) submitted	44.9% ✎	▁▁▁▁▁·▁▃▁▄█▁ beats 22/246	medium · avg 60%	▇ 0.8%	blunt · at its level
Decision Index 0.4.1: Home appliance simulator (case exact accuracy) submitted	0.0% ✎	█▁▁▃▄▄▆▄▁▂▁▃ beats 44/243	medium · avg 37%	▇ 0.6%	sharp · noisy
Decision Index 0.4.1: RouterBench (selected quality (quality objective)) submitted	79.4% ✎	no other model yet		—	failed run: Publisher excludes this protocol because its prompt reveals GPT-4 as the best route. Evidence: https://huggingface.co/spaces/multimodalart/jev-decision-index/blob/c130faf65f82e81d3bdefb01e758c712dc5c5199/data/index.json

retrieval & classification▁▁▁·▁▁▃▁▃▃█▅ #226 of 258 · 4.5% of its score · 5 tests

Trails Bosun v3.1 0.6B LoRA, Crivo-32M v2, Jev-Style 0.8B v3 here, though they rank lower overall.

test	result	vs the others in its run	difficulty	counts	why
Decision Index 0.4.1: BRIGHT (nDCG@10) submitted	33.0% ✎	▁▁▁▁▁▁▁▂▃▄▇█ beats 61/246	medium · avg 38%	▇ 1.7%	blunt · at its level
Decision Index 0.4.1: Amazon ESCI (macro-F1) submitted	29.2% ✎	▁▁▁▁▁▁▃▂▃▃█▅ beats 32/246	medium · avg 47%	▇ 1.7%	at its level
Decision Index 0.4.1: HoVer (accuracy) submitted	57.2% ✎	▆▆▇▅▄▄▃▇▄█▆▄ beats 37/246	easy · avg 74%	▇ 0.6%	blunt · at its level
Decision Index 0.4.1: PhishNChips phishing decisions (accuracy) submitted	51.2% ✎	█▅▅▄▃▂▁▁▁▁·▁ beats 49/245	easy · avg 70%	▇ 0.5%	at its level · noisy
Decision Index 0.4.1: SGD/SGD-X (macro-F1) submitted	58.8% ✎	no other model yet		—	failed run: Publisher reports a builder bug exposing dialogue turns after the classified turn; awaiting corrected rerun. Evidence: https://huggingface.co/spaces/multimodalart/jev-decision-index/blob/c130faf65f82e81d3bdefb01e758c712dc5c5199/data/index.json

language understanding▂▂▂▃▂▅▆▃█▆▄▃ #184 of 256 · 25% of its score · 9 tests

Beats Nari Decision 4B v3.2, Solar Decide Flash, Lev1 3.7B here, though they rank higher overall. Trails DecisionTune 1.0, vera-spark, jdi-ptr14-0.8b here, though they rank lower overall.

test	result	vs the others in its run	difficulty	counts	why
Decision Index 0.4.1: HellaSwag (accuracy) submitted	77.2% ✎	▂▄▂▂▄▇▃▇█▅▃▁ beats 73/246	very easy · avg 85%	▇▇▇▇ 8.9%	sharp · at its level
Decision Index 0.4.1: NLI4CT (macro-F1) submitted	70.8% ✎	▁▁▁▂▂▃▃▄▇▇█▄ beats 81/246	easy · avg 74%	▇▇ 5.6%	at its level
Decision Index 0.4.1: ANLI (macro-F1) submitted	51.9% ✎	▁▂▃▅▅█▂····▁ beats 75/246	medium · avg 60%	▇▇ 4.7%	at its level
Decision Index 0.4.1: iSarcasmEval (Sarcasm F1 · track A, English) submitted	36.1% ✎	▁·▁·▁▁▂▄▅█▇▄ beats 77/246	medium · avg 44%	▇ 1.7%	at its level
Decision Index 0.4.1: ContractNLI (macro-F1) submitted	54.4% ✎	▁▁▁▁▁▁▁▃▂▇▇█ beats 39/246	easy · avg 71%	▇ 1.1%	at its level · noisy
Decision Index 0.4.1: VAST (macro-F1) submitted	43.1% ✎	▁▁▁▂▅▄▄▂▄▆█▄ beats 46/246	medium · avg 62%	▇ 0.9%	at its level
Decision Index 0.4.1: FinEntity (macro-F1) submitted	64.0% ✎	▁···▁▁▁▃▃▆█▃ beats 29/246	easy · avg 84%	▇ 0.9%	noisy
Decision Index 0.4.1: RAGTruth (F1 on hallucinated class) submitted	0.0% ✎	▃▁▁▁▁▁▁▁▁▃▆█ beats 0/246	easy · avg 66%	▇ 0.8%	at its level · noisy

1 more test

test	result	vs the others in its run	difficulty	counts	why
Decision Index 0.4.1: ACOS (per-review F1) submitted	9.1% ✎	▁··▂▃▃▅▇█▇▃▃ beats 48/246	hard · avg 22%	▇ 0.6%	noisy

arts & human taste▁▇▄▃▅▅█▅▄▂▁▁ #156 of 247 · 7.9% of its score · 5 tests

Beats Jegue 0.1 (9B), Standard One 8B SH (v4), Sieve 4B here, though they rank higher overall. Trails Crivo-32M v2, Crivo-68M, Crivo-150M v2 here, though they rank lower overall.

test	result	vs the others in its run	difficulty	counts	why
Decision Index 0.4.1: New Yorker (accuracy) submitted	55.5% ✎	▁▁▂▃▃▃▇▅█▅▃▁ beats 74/246	medium · avg 61%	▇ 3.6%	at its level
Decision Index 0.4.1: BPoMP (accuracy) submitted	81.0% ✎	▁█▄▄▅█▆▅▄▁▁▁ beats 91/246	easy · avg 83%	▇ 1.6%	sharp · at its level · noisy
Decision Index 0.4.1: Humicroedit (accuracy) submitted	58.2% ✎	▁·▁▃▂▂▅█▄▂▁▁ beats 77/246	medium · avg 60%	▇ 1.6%	blunt · at its level
Decision Index 0.4.1: cfcolor (accuracy) submitted	54.2% ✎	▁········▁▇█ beats 50/246	medium · avg 60%	▇ 0.9%	blunt · at its level
Decision Index 0.4.1: Habermas Machine (accuracy) submitted	46.8% ✎	▁▂▄█▇▅▁▁▄▄▁▁ beats 159/246	medium · avg 47%	▇ 0.3%	blunt · at its level

knowledge & reasoning▁▂▂▂▂▃▃▅▅▆█▂ #178 of 263 · 49% of its score · 14 tests

Beats baselabs joey-1 (2B), Aurelis-2B, StartLux-Decision v1.1 0.8B here, though they rank higher overall.

test	result	vs the others in its run	difficulty	counts	why
Decision Index 0.4.1: ARC-Challenge (accuracy) submitted	88.4% ✎	▂▂▂▂▂▂▁▃▄▆█▆ beats 84/246	very easy · avg 90%	▇▇▇ 8.6%	sharp
Decision Index 0.4.1: MMLU (accuracy) submitted	67.3% ✎	▁▄▄▃▃▂▅▇▅█▇▁ beats 88/246	easy · avg 70%	▇▇▇ 7.6%	sharp · at its level
Decision Index 0.4.1: MMLU-Pro (accuracy) submitted	41.7% ✎	▄▄▃▃▂▄▅▇██▅▁ beats 85/246	medium · avg 46%	▇▇▇ 7.4%	sharp · at its level
Decision Index 0.4.1: ARC-Easy (accuracy) submitted	95.8% ✎	▁▁▂▁▂▂▂▂▃▆█▅ beats 88/246	very easy · avg 96%	▇▇▇ 6.4%	sharp · noisy
Decision Index 0.4.1: BBH (accuracy) submitted	51.4% ✎	▄▆▄▄█▇▇▄▃▁▁▁ beats 67/246	easy · avg 66%	▇▇ 6.2%	at its level
Decision Index 0.4.1: WinoGrande (accuracy) submitted	59.5% ✎	█▅█▅▄█▆▄▃▃▁▁ beats 43/246	easy · avg 78%	▇ 3.6%	at its level
Decision Index 0.4.1: GSM8K (accuracy (4- and 10-choice)) submitted	28.8% ✎	▁··▄█▆▆▆▅▅▃▁ beats 73/246	medium · avg 45%	▇ 2.4%	sharp · at its level
Decision Index 0.4.1: GPQA Diamond (accuracy) submitted	30.6% ✎	▂▆▅▅▆█▆▃▁··▁ beats 62/246	medium · avg 39%	▇ 2.3%	blunt · at its level

6 more tests

test	result	vs the others in its run	difficulty	counts	why
Decision Index 0.4.1: MuSR (accuracy) submitted	59.6% ✎	▁▁▂█▅▄▁▁▁▁▁▁ beats 130/246	medium · avg 60%	▇ 1.0%	blunt · at its level
Decision Index 0.4.1: CRUXEval (accuracy) submitted	38.8% ✎	██▄▃▃▄▁▁···▁ beats 46/246	medium · avg 58%	▇ 0.9%	at its level · noisy
Decision Index 0.4.1: CLadder (accuracy) submitted	54.4% ✎	▁███▆▅▃▃▁▁▁▁ beats 38/246	easy · avg 71%	▇ 0.9%	at its level
Decision Index 0.4.1: SATA-Bench (case exact accuracy) submitted	20.8% ✎	▁▁▁▁▁▁▁▂▅█▂▁ beats 80/246	hard · avg 22%	▇ 0.8%	typical
Decision Index 0.4.1: ChessBench (accuracy) submitted	16.9% ✎	▄▇▄▄▄▆▄█▄▁▁▁ beats 131/246	hard · avg 16%	▇ 0.4%	blunt
Decision Index 0.4.1: HLE (accuracy) submitted	12.0% ✎	▁▁▂▃▆█▇▅▅▃▁▁ beats 111/246	very hard · avg 13%	▇ 0.0%	blunt

other tests2 not in a task · 6.6% of its score

test	result	vs the others in its run	difficulty	counts	why
Decision Index 0.4.1: private same skills (equated points)	22.5800 points ✎	▂▂▂▃▃▃▄▅▅▇█▃ beats 44/247	avg 41.206 points	▇▇ 4.3%	blunt
Decision Index 0.4.1: private new domains (equated points)	23.1600 points ✎	▃█▃▃▂▆▃▃▄▄▄▄ beats 80/247	avg 38.861 points	▇ 2.3%	blunt

$ where to run it

Implementation and setup instructions ↗
Published implementation and setup documentation verified. This entry describes a decision readout or evaluated configuration, not necessarily separate trained weights. Follow the referenced version and model settings; hardware compatibility has not been verified. Source · checked 2026-10-31.

Model repository and documentation: github.com/r-ms ↗

$ how the score was worked out · Decision Index 0.4.1 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
Decision Index 0.4.1 · submitted; independent execution unverified	38	▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇ 99%	CC BY 4.0
prior · pull toward the middle		▇ 1%
weighted average		= 41.2

not counted (2)

Decision Index 0.4.1	Decision Index 0.4.1: RouterBench (selected quality (quality objective)) [correct]	79.4%	failed run: Publisher excludes this protocol because its prompt reveals GPT-4 as the best route. Evidence: https://huggingface.co/spaces/multimodalart/jev-decision-index/blob/c130faf65f82e81d3bdefb01e758c712dc5c5199/data/index.json
Decision Index 0.4.1	Decision Index 0.4.1: SGD/SGD-X (macro-F1) [correct]	58.8%	failed run: Publisher reports a builder bug exposing dialogue turns after the classified turn; awaiting corrected rerun. Evidence: https://huggingface.co/spaces/multimodalart/jev-decision-index/blob/c130faf65f82e81d3bdefb01e758c712dc5c5199/data/index.json

$ 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