whatmodel.sh / decision / Jev-Style 0.8B v3

Jev-Style 0.8B v3

+ contribute

Jev-Style 0.8B v3 is an open-weights decision model from chaoliangUNSW, released 24 September 2026. Its published size is 0.87B parameters. Licence: apache-2.0.

Catalogue identity notes

Exact Decision Index tested engine jev-style-0.8b-v3; reported model Jev-Style 0.8B v3. Type: full fine-tune; technique: autoregressive. Comparison entry; no current run command, API price or speed inferred from the benchmark. Current repository metadata checked separately; an unpinned alias is not evidence that today’s weights reproduce this historical run.

Model information sources

Release date ↗ · Weights licence ↗ · Parameter count ↗

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

28.0±1.0#319 of 355 decision modelsfrom 42 benchmark results across 1 source

$ how it compares

The decision models ranked either side of Jev-Style 0.8B v3, out of 357 with a score, then more from chaoliangUNSW.

#	model	score	where	released
319	JEV Ettin 400M Elfsong	29±1	model files
320	ZTC-Judge-4B (DecisionBench b0bb1603) FINAL-Bench	29±2	model files
321	Jev-Style 0.8B v3 chaoliangUNSW	28±1	model files	2026-09-24
322	mxbai-rerank-base-v2 (DecisionBench 3ea9d4df) mixedbread-ai	27±2	model files
323	open-jev-deberta-v3-large (DecisionBench 19bf9a64) com-kotobalabs	27±2	model files
	more from chaoliangUNSW
233	Jev-Style 2B v4 chaoliangUNSW	47±1	model files

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

$ what it's good at

tools & automation▁▃▁▁▃▂▃▄▆██▆ #218 of 250 · 10% of its score · 6 tests

Beats Cam1, DM-JEPA 1.1, Kodiak v0.2 1B (accuracy) here, though they rank higher overall. Trails Prumo (rule-based engine), JevK5 Lite, Crivo-32M v2 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	69.4% ✎	▁▂▁▁▃▂▂▄▅▆█▁ beats 31/246	very easy · avg 92%	▇▇ 6.2%	sharp · at its level · noisy
Decision Index 0.4.1: API-Bank (accuracy) submitted	47.4% ✎	▁▁▁▁▁▁▁▃▃▂█▂ beats 45/236	easy · avg 70%	▇ 1.9%	sharp · at its level · noisy
Decision Index 0.4.1: When2Call (accuracy) submitted	38.8% ✎	▁▁▁▃▃▅▅▆█▇█▃ beats 26/246	easy · avg 72%	▇ 1.0%	at its level · noisy
Decision Index 0.4.1: ToolRet (nDCG@10) submitted	58.3% ✎	▁▁▁▁▁·▁▁▁▄█▁ beats 56/246	medium · avg 60%	▇ 0.7%	blunt · at its level
Decision Index 0.4.1: Home appliance simulator (case exact accuracy) submitted	0.0% ✎	█▁▁▃▄▄▆▄▁▂▁▃ beats 31/243	medium · avg 37%	▇ 0.2%	sharp · too hard for it · noisy
Decision Index 0.4.1: RouterBench (selected quality (quality objective)) submitted	78.4% ✎	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▁▁▁·▁▁▃▁▃▃█▅ #219 of 258 · 4.5% of its score · 7 tests

Beats Intelif Qwen3-4B, mini-jev Qwen3-4B (decision readout), open-jev (pngwn) here, though they rank higher overall. Trails Bosun v3.1 0.6B LoRA, Crivo-32M v2 here, though they rank lower overall.

test	result	vs the others in its run	difficulty	counts	why
Decision Index 0.4.1: Amazon ESCI (macro-F1) submitted	32.0% ✎	▁▁▁▁▁▁▂▃▃▃█▅ beats 39/246	medium · avg 47%	▇ 1.3%	at its level
Decision Index 0.4.1: BRIGHT (nDCG@10) submitted	29.1% ✎	▁▁▁▁▁▁▁▃▂▄▇█ beats 43/246	medium · avg 38%	▇ 1.3%	blunt
Decision Index 0.4.1: CLINC150+OOS (macro-F1) submitted	50.1% ✎	▁▁▁▁▁▁▁▃▃▆█▄ beats 26/238	easy · avg 83%	▇ 0.6%	sharp · at its level · noisy
Decision Index 0.4.1: HoVer (accuracy) submitted	57.1% ✎	▆▆▇▅▄▄▃▇▄█▆▄ beats 36/246	easy · avg 74%	▇ 0.5%	blunt · at its level
Decision Index 0.4.1: PhishNChips phishing decisions (accuracy) submitted	51.2% ✎	█▅▅▄▃▂▁▁▁▁·▁ beats 48/245	easy · avg 70%	▇ 0.5%	at its level · noisy
Decision Index 0.4.1: BANKING77 (macro-F1) submitted	52.4% ✎	▁·▁▁▁▁▁▃▂█▇█ beats 16/238	easy · avg 79%	▇ 0.4%	at its level · noisy
Decision Index 0.4.1: SGD/SGD-X (macro-F1) submitted	59.6% ✎	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▂▃▂▃▂▅▆▃█▆▄▃ #241 of 256 · 23% of its score · 9 tests

Beats Bosun v3.1 1.7B LoRA here, though they rank higher overall. Trails CLM-v0.1-8B Qwen3-8B heads, Qwen-2.5-1B-RLCD (decision readout), MoganAI Lavoir ModernBERT-large 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	35.2% ✎	▂▄▂▂▄▇▃▇█▅▃▁ beats 15/246	very easy · avg 85%	▇▇▇▇ 10%	sharp · at its level
Decision Index 0.4.1: NLI4CT (macro-F1) submitted	57.8% ✎	▁▁▁▂▂▃▃▄▇▇█▄ beats 37/246	easy · avg 74%	▇▇ 5.3%	at its level
Decision Index 0.4.1: ANLI (macro-F1) submitted	28.3% ✎	▁▃▃▅▅█▂····▁ beats 10/246	medium · avg 60%	▇ 3.7%	at its level
Decision Index 0.4.1: iSarcasmEval (Sarcasm F1 · track A, English) submitted	15.6% ✎	▁·▁·▁▁▃▄▅█▇▄ beats 15/246	medium · avg 44%	▇ 1.1%	typical
Decision Index 0.4.1: FinEntity (macro-F1) submitted	69.5% ✎	▁···▁▁▁▁▃▆█▃ beats 40/246	easy · avg 84%	▇ 1.0%	at its level · noisy
Decision Index 0.4.1: ContractNLI (macro-F1) submitted	44.3% ✎	▁▁▁▁▁▁▃▂▂▇▇█ beats 26/246	easy · avg 71%	▇ 1.0%	at its level · noisy
Decision Index 0.4.1: VAST (macro-F1) submitted	33.5% ✎	▁▁▁▃▅▄▄▂▄▆█▄ beats 12/246	medium · avg 62%	▇ 0.7%	at its level
Decision Index 0.4.1: RAGTruth (F1 on hallucinated class) submitted	5.3% ✎	▁▁▁▁▃▁▁▁▁▃▆█ beats 9/246	easy · avg 66%	▇ 0.6%	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	5.6% ✎	▁··▂▃▃▅▇█▇▃▃ beats 29/246	hard · avg 22%	▇ 0.3%	too hard for it · noisy

arts & human taste▁▇▄▃▅▅█▅▄▂▁▁ #239 of 247 · 7.0% of its score · 6 tests

Beats JEV Ettin 150M, JEV FLAN-T5 Large public v2 (342M), Kev 0.8B r15 here, though they rank higher overall. Trails Zev (decision readout), LFM2.5-350M-RLCD, Nirnaya 0.4B 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	37.1% ✎	▁▁▂▃▃▃▇▅█▅▃▁ beats 30/246	medium · avg 61%	▇ 2.8%	at its level
Decision Index 0.4.1: BPoMP (accuracy) submitted	48.8% ✎	▁█▄▄▅█▆▅▄▁▁▁ beats 7/246	easy · avg 83%	▇ 1.8%	sharp · at its level · noisy
Decision Index 0.4.1: Humicroedit (accuracy) submitted	48.9% ✎	▁·▁▃▂▂▅█▄▂▁▁ beats 11/246	medium · avg 60%	▇ 1.4%	blunt · at its level
Decision Index 0.4.1: cfcolor (accuracy) submitted	56.7% ✎	▁········▁▇█ beats 65/246	medium · avg 60%	▇ 0.8%	blunt · at its level
Decision Index 0.4.1: Habermas Machine (accuracy) submitted	32.5% ✎	▁▃▄█▇▅▁▁▄▄▁▁ beats 7/246	medium · avg 47%	▇ 0.2%	blunt · at its level
Decision Index 0.4.1: POP909-CL (accuracy) submitted	15.2% ✎	▄▅▅▃▆▅▃▅▂▅▃█ beats 122/237	hard · avg 21%	▇ 0.1%	sharp · too hard for it · noisy

knowledge & reasoning▁▂▃▂▂▃▃▅▅▆█▂ #229 of 263 · 49% of its score · 14 tests

Beats Decision-1.0-Eos-0.8B (DecisionBench 3c2d6326), Amber 0.4B, vLLM-SR vela3-fast-vision (Vela 3.0, 0.25B) here, though they rank higher overall. Trails laya (DecisionBench 1c5edc17), LFM2.5-2.6B-RLCD here, though they rank lower overall.

test	result	vs the others in its run	difficulty	counts	why
Decision Index 0.4.1: ARC-Easy (accuracy) submitted	75.0% ✎	▁▁▂▁▃▂▂▂▃▆█▅ beats 37/246	very easy · avg 96%	▇▇▇▇▇ 13%	sharp · at its level · noisy
Decision Index 0.4.1: ARC-Challenge (accuracy) submitted	52.8% ✎	▂▂▂▃▂▂▁▃▄▆█▆ beats 31/246	very easy · avg 90%	▇▇▇▇ 11%	sharp · at its level
Decision Index 0.4.1: MMLU (accuracy) submitted	39.0% ✎	▁▄▄▃▃▂▅▇▅█▇▁ beats 31/246	easy · avg 70%	▇▇▇ 6.3%	sharp · at its level
Decision Index 0.4.1: BBH (accuracy) submitted	40.3% ✎	▄▆▄▄█▇▇▄▃▁▁▁ beats 37/246	easy · avg 66%	▇▇ 5.0%	at its level
Decision Index 0.4.1: MMLU-Pro (accuracy) submitted	17.6% ✎	▄▄▃▃▂▄▅▇██▅▁ beats 31/246	medium · avg 46%	▇▇ 4.3%	sharp
Decision Index 0.4.1: WinoGrande (accuracy) submitted	50.4% ✎	█▅█▅▄█▆▄▃▃▁▁ beats 6/246	easy · avg 78%	▇ 3.6%	at its level
Decision Index 0.4.1: GPQA Diamond (accuracy) submitted	26.0% ✎	▂▆▅▅▆█▆▃▁··▁ beats 30/246	medium · avg 39%	▇ 1.7%	blunt · at its level
Decision Index 0.4.1: GSM8K (accuracy (4- and 10-choice)) submitted	19.8% ✎	▁··▄█▆▆▆▅▅▃▁ beats 35/246	medium · avg 45%	▇ 1.3%	sharp

6 more tests

test	result	vs the others in its run	difficulty	counts	why
Decision Index 0.4.1: MuSR (accuracy) submitted	49.3% ✎	▁▁▃█▅▄▁▁▁▁▁▁ beats 27/246	medium · avg 60%	▇ 0.9%	blunt · at its level
Decision Index 0.4.1: CLadder (accuracy) submitted	52.6% ✎	▁███▆▅▃▃▁▁▁▁ beats 25/246	easy · avg 71%	▇ 0.9%	at its level
Decision Index 0.4.1: CRUXEval (accuracy) submitted	40.0% ✎	██▄▃▃▄▁▁···▁ beats 53/246	medium · avg 58%	▇ 0.7%	at its level · noisy
Decision Index 0.4.1: SATA-Bench (case exact accuracy) submitted	3.0% ✎	▁▁▁▃▁▁▁▂▅█▂▁ beats 26/246	hard · avg 22%	▇ 0.4%	typical
Decision Index 0.4.1: ChessBench (accuracy) submitted	10.7% ✎	▄▇▄▄▄▆▄█▄▁▁▁ beats 71/246	hard · avg 16%	▇ 0.3%	blunt
Decision Index 0.4.1: HLE (accuracy) submitted	13.2% ✎	▁▁▂▃▆█▇▅▅▃▁▁ beats 149/246	very hard · avg 13%	▇ 0.0%	blunt

other tests2 not in a task · 5.6% of its score

test	result	vs the others in its run	difficulty	counts	why
Decision Index 0.4.1: private same skills (equated points)	13.7700 points ✎	▂▂▃▃▃▃▄▅▅▇█▃ beats 22/247	avg 41.206 points	▇ 3.6%	blunt
Decision Index 0.4.1: private new domains (equated points)	10.6000 points ✎	▃█▃▃▂▆▃▃▄▄▄▄ beats 21/247	avg 38.861 points	▇ 2.0%	blunt

$ where to run it

Download model files ↗
Published model files verified. This is a verified repository snapshot; the benchmark does not establish an identical weight revision. 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/chaoliangUNSW ↗

$ how the score was worked out · Decision Index 0.4.1 100%

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	42	▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇ 100%	CC BY 4.0
prior · pull toward the middle		▇ 0%
weighted average		= 28.0

not counted (2)

Decision Index 0.4.1	Decision Index 0.4.1: RouterBench (selected quality (quality objective)) [correct]	78.4%	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]	59.6%	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