whatmodel.sh / decision / JoshuaSP diffgemma (decision readout)

JoshuaSP diffgemma (decision readout)

+ contribute

JoshuaSP diffgemma (decision readout) is a source-available decision implementation from JoshuaSP. Its published size is 25.82B parameters.

Catalogue identity notes

Exact Decision Index tested engine joshua-diffusion-full; reported model JoshuaSP diffgemma. Type: inference technique; technique: diffusion. Published checkpoint/base reference: google/diffusiongemma-26B-A4B-it. Implementation: https://github.com/JoshuaSP/open-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.

54.5±1.2#163 of 355 decision modelsfrom 42 benchmark results across 1 source

$ how it compares

The decision models ranked either side of JoshuaSP diffgemma (decision readout), out of 357 with a score.

#	model	score	where	released
163	Winnow-12B EldanRing	55±1	model files
164	ezjev 4B s2 everettjf	55±1	model files
165	JoshuaSP diffgemma (decision readout) JoshuaSP	55±1	source code
166	decisio 0.4.0 (Gemma 4 12B) (decision readout) aminry	54±1	source code	2026-10-04
167	JevK5 9B v0.3.3 alibiserikbay	54±1	model files

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

$ what it's good at

tools & automation▁▃▁▁▁▂▃▄▆██▆ #81 of 250 · 7.5% of its score · 6 tests

Beats Sieve 27B v2, Jev 1.13, Span-02 (Gemma 4 26B-A4B) here, though they rank higher overall. Trails Jev-Omni, baselabs joey-1 (2B), Hopper 12B (trained) here, though they rank lower overall.

test	result	vs the others in its run	difficulty	counts	why
Decision Index 0.4.1: API-Bank (accuracy) submitted	83.7% ✎	▁▁▁▁▁▁▁▂▃▂█▂ beats 138/236	easy · avg 70%	▇ 2.3%	sharp · noisy
Decision Index 0.4.1: BFCL (case exact accuracy) submitted	96.5% ✎	▁▂▁▁▁▂▂▄▅▆█▁ beats 168/246	very easy · avg 92%	▇ 2.0%	sharp · maxed out · noisy
Decision Index 0.4.1: Home appliance simulator (case exact accuracy) submitted	42.0% ✎	█▁▁▃▄▄▆▄▁▂▁▃ beats 118/243	medium · avg 37%	▇ 1.2%	sharp · at its level · noisy
Decision Index 0.4.1: When2Call (accuracy) submitted	75.4% ✎	▁▁▁▂▃▅▅▆█▇█▃ beats 115/246	easy · avg 72%	▇ 1.1%	noisy
Decision Index 0.4.1: ToolRet (nDCG@10) submitted	61.2% ✎	▁▁▁▁▁·▁▁▁▄█▁ beats 77/246	medium · avg 60%	▇ 1.0%	blunt · at its level
Decision Index 0.4.1: RouterBench (selected quality (quality objective)) submitted	79.0% ✎	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▁▁▁·▁▁▁▁▃▃█▅ #119 of 258 · 6.5% of its score · 7 tests

Beats AutoTrust-JEV-27B v1.1, Quyet-1.0-Large, deck31b (Gemma 4 31B) (decision readout) here, though they rank higher overall. Trails vLLM-SR vela3-fast (Vela 3.0, 0.4B), crystal-0.8b, Kas 4B 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	42.9% ✎	▁▁▁▁▁▁▁▂▂▄▇█ beats 126/246	medium · avg 38%	▇ 2.4%	blunt · at its level
Decision Index 0.4.1: Amazon ESCI (macro-F1) submitted	53.4% ✎	▁▁▁▁▁▁▂▂▃▃█▅ beats 135/246	medium · avg 47%	▇ 2.2%	at its level
Decision Index 0.4.1: HoVer (accuracy) submitted	70.9% ✎	▆▆▇▅▄▄▃▇▄█▆▄ beats 103/246	easy · avg 74%	▇ 0.6%	blunt
Decision Index 0.4.1: PhishNChips phishing decisions (accuracy) submitted	85.4% ✎	█▅▅▄▃▂▁▁▁▁·▁ beats 192/245	easy · avg 70%	▇ 0.5%	at its level · noisy
Decision Index 0.4.1: CLINC150+OOS (macro-F1) submitted	83.5% ✎	▁▁▁▁▁▁▁▂▃▆█▄ beats 75/238	easy · avg 83%	▇ 0.4%	sharp · noisy
Decision Index 0.4.1: BANKING77 (macro-F1) submitted	74.3% ✎	▁·▁▁▁▁▁▁▂█▇█ beats 57/238	easy · avg 79%	▇ 0.3%	noisy
Decision Index 0.4.1: SGD/SGD-X (macro-F1) submitted	40.6% ✎	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▂▂▂▃▂▅▆▃█▆▄▃ #152 of 256 · 23% of its score · 9 tests

Trails Decider 2B, Crivo-1B, vera-core-2 here, though they rank lower overall.

test	result	vs the others in its run	difficulty	counts	why
Decision Index 0.4.1: NLI4CT (macro-F1) submitted	78.4% ✎	▁▁▁▂▂▃▃▄▇▇█▄ beats 135/246	easy · avg 74%	▇▇ 5.4%	typical
Decision Index 0.4.1: HellaSwag (accuracy) submitted	83.3% ✎	▂▄▂▂▄▇▃▇█▅▃▁ beats 101/246	very easy · avg 85%	▇▇ 5.3%	sharp
Decision Index 0.4.1: ANLI (macro-F1) submitted	66.4% ✎	▁▂▃▅▅█▂····▁ beats 134/246	medium · avg 60%	▇▇ 5.3%	at its level
Decision Index 0.4.1: iSarcasmEval (Sarcasm F1 · track A, English) submitted	66.0% ✎	▁·▁·▁▁▂▄▅█▇▄ beats 211/246	medium · avg 44%	▇ 2.3%	at its level
Decision Index 0.4.1: ACOS (per-review F1) submitted	24.5% ✎	▁··▂▃▃▅▇█▇▃▃ beats 141/246	hard · avg 22%	▇ 1.1%	at its level · noisy
Decision Index 0.4.1: ContractNLI (macro-F1) submitted	76.0% ✎	▁▁▁▁▁▁▁▂▂▇▇█ beats 126/246	easy · avg 71%	▇ 1.1%	at its level · noisy
Decision Index 0.4.1: VAST (macro-F1) submitted	55.7% ✎	▁▁▁▂▅▄▄▂▄▆█▄ beats 97/246	medium · avg 62%	▇ 1.0%	at its level
Decision Index 0.4.1: RAGTruth (F1 on hallucinated class) submitted	70.4% ✎	▁▁▁▁▁▁▁▁▁▃▆█ beats 98/246	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: FinEntity (macro-F1) submitted	89.0% ✎	▁···▁▁▁▁▃▆█▃ beats 139/246	easy · avg 84%	▇ 0.7%	noisy

arts & human taste▁▇▄▃▅▅█▅▄▂▁▁ #131 of 247 · 9.3% of its score · 6 tests

Beats OneJev1.6-4B, Nace.AI Drex v1.5, Cerno 4B v1.1 here, though they rank higher overall. Trails vLLM-SR vela3-fast-vision (Vela 3.0, 0.25B), Crivo-400M, 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: New Yorker (accuracy) submitted	63.6% ✎	▁▁▂▃▃▃▇▅█▅▃▁ beats 122/246	medium · avg 61%	▇▇ 4.0%	at its level
Decision Index 0.4.1: Humicroedit (accuracy) submitted	63.0% ✎	▁·▁▃▂▂▅█▄▂▁▁ beats 171/246	medium · avg 60%	▇ 1.9%	blunt · at its level
Decision Index 0.4.1: BPoMP (accuracy) submitted	86.9% ✎	▁█▄▄▅█▆▅▄▁▁▁ beats 116/246	easy · avg 83%	▇ 1.2%	sharp · noisy
Decision Index 0.4.1: cfcolor (accuracy) submitted	58.2% ✎	▁········▁▇█ beats 81/246	medium · avg 60%	▇ 1.1%	blunt · at its level
Decision Index 0.4.1: POP909-CL (accuracy) submitted	2.5% ✎	▄▅▅▃▆▅▃▅▂▅▃█ beats 56/237	hard · avg 21%	▇ 0.9%	sharp · at its level · noisy
Decision Index 0.4.1: Habermas Machine (accuracy) submitted	45.0% ✎	▁▂▄█▇▅▁▁▄▄▁▁ beats 136/246	medium · avg 47%	▇ 0.3%	blunt · at its level

knowledge & reasoning▁▂▂▂▂▃▃▅▅▆█▂ #113 of 263 · 45% of its score · 14 tests

Beats OneJev1.6-4B, Seb-9B v2, Mica v0.5 4B here, though they rank higher overall. Trails PurpleMIST-Flash-1.0, Jevfire Qwen3.8-27B (decision readout), JevK5 9B v0.3.3 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	56.9% ✎	▄▄▃▃▂▄▅▇██▅▁ beats 143/246	medium · avg 46%	▇▇▇▇ 10%	sharp · at its level
Decision Index 0.4.1: MMLU (accuracy) submitted	79.3% ✎	▁▄▄▃▃▂▅▇▅█▇▁ beats 148/246	easy · avg 70%	▇▇▇ 7.1%	sharp
Decision Index 0.4.1: BBH (accuracy) submitted	70.7% ✎	▄▆▄▄█▇▇▄▃▁▁▁ beats 137/246	easy · avg 66%	▇▇▇ 6.4%	at its level
Decision Index 0.4.1: ARC-Challenge (accuracy) submitted	94.5% ✎	▂▂▂▂▂▂▁▃▄▆█▆ beats 136/246	very easy · avg 90%	▇▇ 4.1%	sharp · maxed out
Decision Index 0.4.1: GSM8K (accuracy (4- and 10-choice)) submitted	33.1% ✎	▁··▄█▆▆▆▅▅▃▁ beats 85/246	medium · avg 45%	▇ 3.4%	sharp · at its level
Decision Index 0.4.1: WinoGrande (accuracy) submitted	73.6% ✎	█▅█▅▄█▆▄▃▃▁▁ beats 101/246	easy · avg 78%	▇ 3.2%	typical
Decision Index 0.4.1: GPQA Diamond (accuracy) submitted	44.9% ✎	▂▆▅▅▆█▆▃▁··▁ beats 152/246	medium · avg 39%	▇ 3.1%	blunt · at its level
Decision Index 0.4.1: ARC-Easy (accuracy) submitted	98.2% ✎	▁▁▂▁▂▂▂▂▃▆█▅ beats 143/246	very easy · avg 96%	▇ 2.1%	sharp · maxed out · noisy

6 more tests

test	result	vs the others in its run	difficulty	counts	why
Decision Index 0.4.1: SATA-Bench (case exact accuracy) submitted	27.5% ✎	▁▁▁▁▁▁▁▂▅█▂▁ beats 120/246	hard · avg 22%	▇ 1.4%	at its level
Decision Index 0.4.1: MuSR (accuracy) submitted	61.2% ✎	▁▁▂█▅▄▁▁▁▁▁▁ beats 145/246	medium · avg 60%	▇ 1.2%	blunt · at its level
Decision Index 0.4.1: CRUXEval (accuracy) submitted	64.7% ✎	██▄▃▃▄▁▁···▁ beats 154/246	medium · avg 58%	▇ 1.1%	at its level · noisy
Decision Index 0.4.1: CLadder (accuracy) submitted	67.8% ✎	▁███▆▅▃▃▁▁▁▁ beats 118/246	easy · avg 71%	▇ 1.0%	at its level
Decision Index 0.4.1: ChessBench (accuracy) submitted	14.2% ✎	▄▇▄▄▄▆▄█▄▁▁▁ beats 108/246	hard · avg 16%	▇ 0.6%	blunt
Decision Index 0.4.1: HLE (accuracy) submitted	9.8% ✎	▁▁▂▃▆█▇▅▅▃▁▁ beats 26/246	very hard · avg 13%	▇ 0.0%	blunt

other tests2 not in a task · 8.2% of its score

test	result	vs the others in its run	difficulty	counts	why
Decision Index 0.4.1: private same skills (equated points)	49.5400 points ✎	▂▂▂▃▃▃▄▅▅▇█▃ beats 142/247	avg 41.206 points	▇▇ 5.3%	blunt
Decision Index 0.4.1: private new domains (equated points)	49.5000 points ✎	▃█▃▃▂▆▃▃▄▄▄▄ beats 155/247	avg 38.861 points	▇ 2.9%	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/JoshuaSP ↗

$ 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	42	▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇ 99%	CC BY 4.0
prior · pull toward the middle		▇ 1%
weighted average		= 54.5

not counted (2)

Decision Index 0.4.1	Decision Index 0.4.1: RouterBench (selected quality (quality objective)) [correct]	79.0%	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]	40.6%	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