whatmodel.sh / decision / JPT-4B Qwen3.5-4B LoRA

JPT-4B Qwen3.5-4B LoRA

+ contribute

JPT-4B Qwen3.5-4B LoRA is an open-weights decision model from kirp. Its published size is 4.66B parameters. Licence: cc-by-nc-4.0.

Catalogue identity notes

Exact Decision Index tested engine jpt-4b; reported model JPT-4B Qwen3.5-4B LoRA. Type: LoRA; technique: autoregressive. Published checkpoint/base reference: 277ae37b7a6398a8ec3b86712ac16c1991ab6ef1. Implementation: https://github.com/tic-top/llm2jev. 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

· Weights licence ↗ · Parameter count ↗

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

50.4±1.1#202 of 355 decision modelsfrom 56 benchmark results across 3 sources

$ how it compares

The decision models ranked either side of JPT-4B Qwen3.5-4B LoRA, out of 357 with a score, then more from kirp.

#	model	score	where	RAM
202	Together Tev1 Qwen3.5-4B Together	51±1	model files
203	OneJev-4B (JevBench mini) OmniJev	51±2	source code
204	JPT-4B Qwen3.5-4B LoRA kirp	50±1	model files
205	Bespoke Nimble 9B v2 Bespoke Labs	50±1	local	~20GB
206	JevK5 v0.3 alibiserikbay	50±1	model files
	more from kirp
128	JPT-35B-A3B kirp	58±1	model files
178	JPT-9B Qwen3.5-9B LoRA kirp	54±1	model files
289	JPT-0.8B Qwen3.5-0.8B LoRA kirp	33±1	model files

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

$ what it's good at

tools & automation▁▃▁▁▁▂▃▄▆██▆ #165 of 250 · 7.1% of its score · 6 tests

Beats LiquidAI d1-3B, vLLM-SR Decision 2.0 Nox 4B, Decider 4B here, though they rank higher overall. Trails JevK5 Lite, Gevva E2B, Amber 0.4B 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	91.6% ✎	▁▂▁▁▁▂▂▄▅▆█▁ beats 85/246	very easy · avg 92%	▇ 2.2%	sharp · maxed out · noisy
Decision Index 0.4.1: API-Bank (accuracy) submitted	68.7% ✎	▁▁▁▁▁▁▁▂▃▃█▂ beats 87/236	easy · avg 70%	▇ 2.2%	sharp · at its level · noisy
Decision Index 0.4.1: When2Call (accuracy) submitted	64.0% ✎	▁▁▁▂▃▅▅▆█▇█▃ beats 76/246	easy · avg 72%	▇ 1.0%	at its level · noisy
Decision Index 0.4.1: Home appliance simulator (case exact accuracy) submitted	12.5% ✎	█▁▁▃▄▄▆▄▁▂▁▃ beats 76/243	medium · avg 37%	▇ 0.9%	sharp · at its level · noisy
Decision Index 0.4.1: ToolRet (nDCG@10) submitted	62.1% ✎	▁▁▁▁▁·▁▁▁▄█▁ beats 90/246	medium · avg 60%	▇ 0.8%	blunt · at its level
Decision Index 0.4.1: RouterBench (selected quality (quality objective)) submitted	79.9% ✎	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▁▁▁·▁▁▁▁▃▃█▅ #169 of 258 · 5.4% of its score · 7 tests

Beats Solar Decide, decisio 0.4.0 (Qwen3.6-35B-A3B) (decision readout), JevK5 9B v0.3.3 here, though they rank higher overall. Trails Bosun v3.1 1.7B LoRA, Dinah-1, DecisionTune 1.0 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	42.1% ✎	▁▁▁▁▁▁▁▂▂▄▇█ beats 119/246	medium · avg 38%	▇ 1.9%	blunt · at its level
Decision Index 0.4.1: Amazon ESCI (macro-F1) submitted	46.4% ✎	▁▁▁▁▁▁▂▂▃▃█▅ beats 89/246	medium · avg 47%	▇ 1.8%	at its level
Decision Index 0.4.1: HoVer (accuracy) submitted	60.9% ✎	▆▆▇▅▄▄▃▇▄█▆▄ beats 55/246	easy · avg 74%	▇ 0.5%	blunt · at its level
Decision Index 0.4.1: PhishNChips phishing decisions (accuracy) submitted	68.9% ✎	█▅▅▄▃▂▁▁▁▁·▁ beats 125/245	easy · avg 70%	▇ 0.5%	at its level · noisy
Decision Index 0.4.1: CLINC150+OOS (macro-F1) submitted	64.5% ✎	▁▁▁▁▁▁▁▃▃▆█▄ beats 35/238	easy · avg 83%	▇ 0.4%	sharp · noisy
Decision Index 0.4.1: BANKING77 (macro-F1) submitted	72.1% ✎	▁·▁▁▁▁▁▁▂█▇█ beats 50/238	easy · avg 79%	▇ 0.3%	noisy
Decision Index 0.4.1: SGD/SGD-X (macro-F1) submitted	60.5% ✎	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▂▂▂▃▂▅▆▃█▆▄▃ #167 of 256 · 21% of its score · 9 tests

Beats PurpleMIST-Flash-1.0, Hopper 12B (trained) here, though they rank higher overall. Trails DecisionTune 1.0, Decider 2B, vLLM-SR d3-lite (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	81.0% ✎	▂▄▂▂▄▇▃▇█▅▃▁ beats 86/246	very easy · avg 85%	▇▇ 5.7%	sharp
Decision Index 0.4.1: NLI4CT (macro-F1) submitted	78.5% ✎	▁▁▁▂▂▃▃▄▇▇█▄ beats 136/246	easy · avg 74%	▇▇ 4.8%	at its level
Decision Index 0.4.1: ANLI (macro-F1) submitted	62.0% ✎	▁▂▃▅▅█▂····▁ beats 111/246	medium · avg 60%	▇▇ 4.5%	at its level
Decision Index 0.4.1: iSarcasmEval (Sarcasm F1 · track A, English) submitted	51.7% ✎	▁·▁·▁▁▂▄▅█▇▄ beats 144/246	medium · avg 44%	▇ 1.9%	at its level
Decision Index 0.4.1: ContractNLI (macro-F1) submitted	79.7% ✎	▁▁▁▁▁▁▁▂▂▇▇█ beats 159/246	easy · avg 71%	▇ 0.9%	at its level · noisy
Decision Index 0.4.1: VAST (macro-F1) submitted	63.4% ✎	▁▁▁▂▅▄▄▃▄▆█▄ beats 115/246	medium · avg 62%	▇ 0.9%	at its level
Decision Index 0.4.1: ACOS (per-review F1) submitted	20.0% ✎	▁··▂▃▃▅▇█▇▃▃ beats 109/246	hard · avg 22%	▇ 0.8%	noisy
Decision Index 0.4.1: RAGTruth (F1 on hallucinated class) submitted	74.2% ✎	▁▁▁▁▁▁▁▁▁▃▆█ beats 126/246	easy · avg 66%	▇ 0.7%	at its level · noisy

1 more test

test	result	vs the others in its run	difficulty	counts	why
Decision Index 0.4.1: FinEntity (macro-F1) submitted	91.9% ✎	▁···▁▁▁▁▃▆█▃ beats 178/246	easy · avg 84%	▇ 0.6%	noisy

arts & human taste▁▇▄▃▅▅█▅▄▂▁▁ #160 of 247 · 7.9% of its score · 6 tests

Beats Jegue 0.1 (9B), Standard One 8B SH (v4), Saif 4B here, though they rank higher overall. Trails Crivo-17M, Crivo-32M v2, Crivo-68M 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	57.4% ✎	▁▁▂▃▃▃▇▅█▅▃▁ beats 84/246	medium · avg 61%	▇ 3.4%	at its level
Decision Index 0.4.1: Humicroedit (accuracy) submitted	59.7% ✎	▁·▁▃▂▂▅█▄▂▁▁ beats 90/246	medium · avg 60%	▇ 1.6%	blunt · at its level
Decision Index 0.4.1: BPoMP (accuracy) submitted	80.5% ✎	▁█▄▄▅█▆▅▄▁▁▁ beats 88/246	easy · avg 83%	▇ 1.2%	sharp · noisy
Decision Index 0.4.1: cfcolor (accuracy) submitted	60.0% ✎	▁········▁▇█ beats 100/246	medium · avg 60%	▇ 0.9%	blunt · at its level
Decision Index 0.4.1: POP909-CL (accuracy) submitted	1.7% ✎	▄▅▅▃▆▅▃▅▂▅▃█ beats 39/237	hard · avg 21%	▇ 0.6%	sharp · noisy
Decision Index 0.4.1: Habermas Machine (accuracy) submitted	38.9% ✎	▁▂▄█▇▅▁▁▄▄▁▁ beats 48/246	medium · avg 47%	▇ 0.3%	blunt · at its level

knowledge & reasoning▁▂▂▂▂▃▃▅▅▆█▂ #149 of 263 · 40% of its score · 14 tests

Beats Winnow-12B (DecisionBench b6ac22b0), StartLux-Decision v1.1 2B, Solomon v1.1 here, though they rank higher overall. Trails Ryotide Qwen v0.3.1 (decision readout), razorback16 diffgemma (decision readout), Solar Decide Flash here, though they rank lower overall.

test	result	vs the others in its run	difficulty	counts	why
Decision Index 0.4.1: MMLU-Pro (accuracy) submitted	48.9% ✎	▄▄▃▃▂▄▅▇██▅▁ beats 106/246	medium · avg 46%	▇▇▇ 8.4%	sharp · at its level
Decision Index 0.4.1: MMLU (accuracy) submitted	73.8% ✎	▁▄▄▃▃▂▅▇▅█▇▁ beats 113/246	easy · avg 70%	▇▇▇ 6.5%	sharp · at its level
Decision Index 0.4.1: BBH (accuracy) submitted	63.1% ✎	▄▆▄▄█▇▇▄▃▁▁▁ beats 97/246	easy · avg 66%	▇▇ 5.7%	at its level
Decision Index 0.4.1: ARC-Challenge (accuracy) submitted	92.8% ✎	▂▂▂▂▂▂▁▃▄▆█▆ beats 110/246	very easy · avg 90%	▇▇ 4.7%	sharp
Decision Index 0.4.1: WinoGrande (accuracy) submitted	71.3% ✎	█▅█▅▄█▆▄▃▃▁▁ beats 94/246	easy · avg 78%	▇ 2.9%	typical
Decision Index 0.4.1: GSM8K (accuracy (4- and 10-choice)) submitted	36.3% ✎	▁··▄█▆▆▆▅▅▃▁ beats 97/246	medium · avg 45%	▇ 2.8%	sharp · at its level
Decision Index 0.4.1: ARC-Easy (accuracy) submitted	97.8% ✎	▁▁▂▁▂▂▂▂▃▆█▅ beats 114/246	very easy · avg 96%	▇ 2.6%	sharp · maxed out · noisy
Decision Index 0.4.1: GPQA Diamond (accuracy) submitted	42.9% ✎	▂▆▅▅▆█▆▃▁··▁ beats 137/246	medium · avg 39%	▇ 2.5%	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	55.6% ✎	▁▁▂█▅▄▁▁▁▁▁▁ beats 76/246	medium · avg 60%	▇ 1.0%	blunt · at its level
Decision Index 0.4.1: SATA-Bench (case exact accuracy) submitted	21.8% ✎	▁▁▁▁▁▁▁▂▅█▂▁ beats 82/246	hard · avg 22%	▇ 1.0%	typical
Decision Index 0.4.1: CRUXEval (accuracy) submitted	50.9% ✎	██▄▃▃▄▁▁···▁ beats 115/246	medium · avg 58%	▇ 0.9%	at its level · noisy
Decision Index 0.4.1: CLadder (accuracy) submitted	65.0% ✎	▁███▆▅▃▃▁▁▁▁ beats 96/246	easy · avg 71%	▇ 0.8%	at its level
Decision Index 0.4.1: ChessBench (accuracy) submitted	13.8% ✎	▄▇▄▄▄▆▄█▄▁▁▁ beats 101/246	hard · avg 16%	▇ 0.5%	blunt
Decision Index 0.4.1: HLE (accuracy) submitted	11.4% ✎	▁▁▂▃▆█▇▅▅▃▁▁ beats 83/246	very hard · avg 13%	▇ 0.0%	blunt

vision▁·▃▂▃▃▇▄▄█▄▃ #49 of 79 · 11% of its score · 14 tests

Beats StartLux-Decision v1.1 2B, Winnow-12B, RSI-Jev v6.1 VL 4B here, though they rank higher overall.

test	result	vs the others in its run	difficulty	counts	why
Decision Index Vision 0.4 · Mind2Web (chance-corrected points) submitted	66.7000 points ✎	▁·▁▁▃▁▄▄▃▅█▂ beats 23/61	avg 63.423 points	▇ 1.3%	blunt
Decision Index Vision 0.4 · MMMU-Pro vision (chance-corrected points) submitted	29.0000 points ✎	▁▁▅▁▄▁▄█▄▃▇▇ beats 24/61	avg 22.797 points	▇ 1.2%	blunt
Decision Index Vision 0.4 · Winoground (chance-corrected points) submitted	77.0000 points ✎	▁·▁▁▁▁·▂▁▃▅█ beats 24/61	avg 66.987 points	▇ 1.1%	blunt
Decision Index Vision 0.4: private same skills (points)	53.4400 points ✎	▁▁▃▃▁▂▄▆▃▃▅█ beats 28/61	avg 48.067 points	▇ 1.1%	blunt
Decision Index Vision 0.4 · Moderation (Hateful Memes) (chance-corrected points) submitted	26.3000 points ✎	▆▃▃▆▄▄▇██▁▆▆ beats 18/61	avg 28.443 points	▇ 0.9%	blunt
Decision Index Vision 0.4 · RealWorldQA (chance-corrected points) submitted	62.5400 points ✎	▁·▂▂▁▃▃▅▅▃▂█ beats 36/61	avg 53.651 points	▇ 0.8%	blunt
Decision Index Vision 0.4 · CharXiv (chance-corrected points) submitted	69.6700 points ✎	▂▁▁▁▁▃▁▅▅▆██ beats 27/61	avg 62.137 points	▇ 0.8%	blunt
Benchmark Heaven imagejevbench v0.3.0: intelligence (points)	38.2663 points ✎	▃·▃▃··▅█▅▅·▃ beats 6/12	avg 32.235 points	▇ 0.7%	blunt

6 more tests

test	result	vs the others in its run	difficulty	counts	why
Decision Index Vision 0.4 · InfographicVQA (chance-corrected points) submitted	90.7100 points ✎	▁·▁▂▁▁▃▁▂▃▅█ beats 28/61	avg 83.859 points	▇ 0.7%	blunt
Decision Index Vision 0.4 · R-Bench-M (chance-corrected points) submitted	19.5100 points ✎	▂▄▇▆█▄▄▇▂··▁ beats 23/61	avg 19.676 points	▇ 0.6%	blunt
Decision Index Vision 0.4 · BLINK (chance-corrected points) submitted	53.7700 points ✎	▁···▁▂▂▂█▄▂▃ beats 34/61	avg 44.994 points	▇ 0.6%	blunt
Decision Index Vision 0.4 · CV-Bench (chance-corrected points) submitted	75.8800 points ✎	▁▁▁▁▂▁▁▁▄█▃▃ beats 36/61	avg 67.662 points	▇ 0.5%	blunt
Decision Index Vision 0.4 · KIE (CORD+FUNSD) (chance-corrected points) submitted	90.4500 points ✎	▁··▁▁▁▁▁▂▄█▁ beats 24/61	avg 85.184 points	▇ 0.4%	blunt
Benchmark Heaven imagejevbench v0.3.0: calibration (points)	72.9769 points ✎	▃···▃·▅█▅▃▃▅ beats 7/12	avg 68.584 points	▇ 0.3%	blunt

other tests2 not in a task · 6.7% of its score

test	result	vs the others in its run	difficulty	counts	why
Decision Index 0.4.1: private same skills (equated points)	39.5200 points ✎	▂▂▂▃▃▃▄▅▅▇█▃ beats 96/247	avg 41.206 points	▇▇ 4.3%	blunt
Decision Index 0.4.1: private new domains (equated points)	39.6100 points ✎	▃█▃▃▂▆▃▃▄▄▄▄ beats 126/247	avg 38.861 points	▇ 2.4%	blunt

$ 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/kirp ↗

$ how the score was worked out · Decision Index 0.4.1 89%, Decision Index Vision 0.4 10%, Benchmark Heaven (own decisio… 1%

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	▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇ 89%	CC BY 4.0
Decision Index Vision 0.4 · submitted; independent execution unverified	12	▇▇▇ 10%	CC BY 4.0
Benchmark Heaven (own decision evaluations)	2	▇ 1%	CC BY 4.0
prior · pull toward the middle		▇ 1%
weighted average		= 50.4

not counted (2)

Decision Index 0.4.1	Decision Index 0.4.1: RouterBench (selected quality (quality objective)) [correct]	79.9%	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]	60.5%	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