whatmodel.sh / decision / SemIf Qwen3.5-4B Base (decision readout)

SemIf Qwen3.5-4B Base (decision readout)

+ contribute

SemIf Qwen3.5-4B Base (decision readout) is a source-available decision implementation from TheoLeeCJ. Its published size is 4.66B parameters.

Catalogue identity notes

Exact Decision Index tested engine semif; reported model SemIf Qwen3.5-4B Base. Type: inference technique; technique: autoregressive. Implementation: https://github.com/TheoLeeCJ/openjev. 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.

45.5±1.2#240 of 355 decision modelsfrom 38 benchmark results across 1 source

$ how it compares

The decision models ranked either side of SemIf Qwen3.5-4B Base (decision readout), out of 357 with a score, then more from TheoLeeCJ.

#	model	score	where	released
240	Kev-4B (JevBench mini) Jared Palmer	46±2	source code
241	TokenRhythm NeoHorse-Jev-4B TokenRhythm	46±1	model files	2026-09-25
242	SemIf Qwen3.5-4B Base (decision readout) TheoLeeCJ	46±1	source code
243	RSI-Jev v3.0 2B (JevBench mini) Shanghua Gao	45±2	source code
244	Metask-Jev-4B Metask-AI-Lab	45±1	model files
	more from TheoLeeCJ
180	SemIf (DecisionBench 1f2dea3e) TheoLeeCJ	54±2	source code

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

$ what it's good at

tools & automation▁▃▁▁▁▂▃▄▆██▆ #137 of 250 · 6.3% of its score · 5 tests

Beats Cerno 4B v1.1, Trio-Spark DI (9B), Decider 35B-A3B here, though they rank higher overall. Trails Amber 0.4B, OneJev1.6-0.8B 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	93.3% ✎	▁▂▁▁▁▂▂▄▅▆█▁ beats 111/246	very easy · avg 92%	▇ 3.4%	sharp · noisy
Decision Index 0.4.1: When2Call (accuracy) submitted	54.7% ✎	▁▁▁▂▃▅▅▆█▇█▃ beats 54/246	easy · avg 72%	▇ 1.2%	at its level · noisy
Decision Index 0.4.1: ToolRet (nDCG@10) submitted	59.6% ✎	▁▁▁▁▁·▁▁▁▄█▁ beats 61/246	medium · avg 60%	▇ 0.9%	blunt · at its level
Decision Index 0.4.1: Home appliance simulator (case exact accuracy) submitted	9.1% ✎	█▁▁▃▄▄▆▄▁▂▁▃ beats 67/243	medium · avg 37%	▇ 0.8%	sharp · at its level · noisy
Decision Index 0.4.1: RouterBench (selected quality (quality objective)) submitted	79.5% ✎	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▁▁▁·▁▁▁▃▃▃█▅ #208 of 258 · 5.0% of its score · 5 tests

Trails Bosun v3.1 0.6B LoRA, Mola-68M, Bosun v3.1 1.7B LoRA 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	41.4% ✎	▁▁▁▁▁▁▁▂▂▄▇█ beats 110/246	medium · avg 38%	▇ 1.9%	blunt · at its level
Decision Index 0.4.1: Amazon ESCI (macro-F1) submitted	36.2% ✎	▁▁▁▁▁▁▂▂▃▃█▅ beats 50/246	medium · avg 47%	▇ 1.9%	at its level
Decision Index 0.4.1: HoVer (accuracy) submitted	61.6% ✎	▆▆▇▅▄▄▃▇▄█▆▄ beats 60/246	easy · avg 74%	▇ 0.6%	blunt · at its level
Decision Index 0.4.1: PhishNChips phishing decisions (accuracy) submitted	60.1% ✎	█▅▅▄▃▂▁▁▁▁·▁ beats 87/245	easy · avg 70%	▇ 0.5%	at its level · noisy
Decision Index 0.4.1: SGD/SGD-X (macro-F1) submitted	56.8% ✎	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▂▂▂▃▂▅▆▃█▆▄▃ #159 of 256 · 25% of its score · 9 tests

Beats PurpleMIST-Flash-1.0, Hopper 12B (trained), Jev-Omni here, though they rank higher overall. Trails DecisionTune 1.0, Decider 2B, Crivo-1B 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	82.7% ✎	▂▄▂▂▄▇▃▇█▅▃▁ beats 98/246	very easy · avg 85%	▇▇▇ 7.9%	sharp
Decision Index 0.4.1: NLI4CT (macro-F1) submitted	72.0% ✎	▁▁▁▂▂▃▃▄▇▇█▄ beats 86/246	easy · avg 74%	▇▇ 5.6%	at its level
Decision Index 0.4.1: ANLI (macro-F1) submitted	54.9% ✎	▁▂▃▅▅█▂····▁ beats 82/246	medium · avg 60%	▇▇ 5.0%	at its level
Decision Index 0.4.1: iSarcasmEval (Sarcasm F1 · track A, English) submitted	39.0% ✎	▁·▁·▁▁▂▄▅█▇▄ beats 88/246	medium · avg 44%	▇ 1.9%	at its level
Decision Index 0.4.1: ContractNLI (macro-F1) submitted	59.5% ✎	▁▁▁▁▁▁▁▃▂▇▇█ beats 49/246	easy · avg 71%	▇ 1.1%	at its level · noisy
Decision Index 0.4.1: VAST (macro-F1) submitted	44.0% ✎	▁▁▁▂▅▄▄▂▄▆█▄ beats 54/246	medium · avg 62%	▇ 1.0%	at its level
Decision Index 0.4.1: FinEntity (macro-F1) submitted	66.0% ✎	▁···▁▁▁▁▃▆█▃ beats 34/246	easy · avg 84%	▇ 0.8%	noisy
Decision Index 0.4.1: RAGTruth (F1 on hallucinated class) submitted	53.4% ✎	▁▁▁▁▁▁▁▁▁▃▆█ beats 63/246	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	13.5% ✎	▁··▂▃▃▅▇█▇▃▃ beats 75/246	hard · avg 22%	▇ 0.8%	noisy

arts & human taste▁▇▄▃▅▅█▅▄▂▁▁ #139 of 247 · 8.3% of its score · 5 tests

Beats OneJev1.6-4B, Nace.AI Drex v1.5, Jegue 0.1 (9B) here, though they rank higher overall. Trails Crivo-68M, Crivo-150M v2, vLLM-SR vela3-fast-vision (Vela 3.0, 0.25B) 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	57.8% ✎	▁▁▂▃▃▃▇▅█▅▃▁ beats 87/246	medium · avg 61%	▇▇ 3.8%	at its level
Decision Index 0.4.1: Humicroedit (accuracy) submitted	59.9% ✎	▁·▁▃▂▂▅█▄▂▁▁ beats 95/246	medium · avg 60%	▇ 1.7%	blunt · at its level
Decision Index 0.4.1: BPoMP (accuracy) submitted	85.2% ✎	▁█▄▄▅█▆▅▄▁▁▁ beats 109/246	easy · avg 83%	▇ 1.5%	sharp · noisy
Decision Index 0.4.1: cfcolor (accuracy) submitted	56.6% ✎	▁········▁▇█ beats 62/246	medium · avg 60%	▇ 1.0%	blunt · at its level
Decision Index 0.4.1: Habermas Machine (accuracy) submitted	44.3% ✎	▁▂▄█▇▅▁▁▄▄▁▁ beats 125/246	medium · avg 47%	▇ 0.3%	blunt · at its level

knowledge & reasoning▁▂▂▂▂▃▃▅▅▆█▂ #161 of 263 · 48% of its score · 14 tests

Beats StartLux-Decision v1.1 2B, Solomon v1.1, Standard One 3B SH (v4) here, though they rank higher overall.

test	result	vs the others in its run	difficulty	counts	why
Decision Index 0.4.1: MMLU-Pro (accuracy) submitted	43.7% ✎	▄▄▃▃▂▄▅▇██▅▁ beats 93/246	medium · avg 46%	▇▇▇ 8.6%	sharp · at its level
Decision Index 0.4.1: MMLU (accuracy) submitted	71.4% ✎	▁▄▄▃▃▂▅▇▅█▇▁ beats 103/246	easy · avg 70%	▇▇▇ 7.7%	sharp · at its level
Decision Index 0.4.1: ARC-Challenge (accuracy) submitted	91.8% ✎	▂▂▂▂▂▂▁▃▄▆█▆ beats 105/246	very easy · avg 90%	▇▇▇ 7.1%	sharp
Decision Index 0.4.1: BBH (accuracy) submitted	57.6% ✎	▄▆▄▄█▇▇▄▃▁▁▁ beats 78/246	easy · avg 66%	▇▇▇ 6.5%	at its level
Decision Index 0.4.1: ARC-Easy (accuracy) submitted	96.8% ✎	▁▁▂▁▂▂▂▂▃▆█▅ beats 101/246	very easy · avg 96%	▇▇ 4.6%	sharp · maxed out · noisy
Decision Index 0.4.1: WinoGrande (accuracy) submitted	66.0% ✎	█▅█▅▄█▆▄▃▃▁▁ beats 60/246	easy · avg 78%	▇ 3.5%	at its level
Decision Index 0.4.1: GSM8K (accuracy (4- and 10-choice)) submitted	24.5% ✎	▁··▄█▆▆▆▅▅▃▁ beats 54/246	medium · avg 45%	▇ 2.9%	sharp · at its level
Decision Index 0.4.1: GPQA Diamond (accuracy) submitted	36.7% ✎	▂▆▅▅▆█▆▃▁··▁ beats 93/246	medium · avg 39%	▇ 2.6%	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	54.6% ✎	▁▁▂█▅▄▁▁▁▁▁▁ beats 61/246	medium · avg 60%	▇ 1.1%	blunt · at its level
Decision Index 0.4.1: CRUXEval (accuracy) submitted	39.0% ✎	██▄▃▃▄▁▁···▁ beats 48/246	medium · avg 58%	▇ 1.0%	at its level · noisy
Decision Index 0.4.1: CLadder (accuracy) submitted	56.0% ✎	▁███▆▅▃▃▁▁▁▁ beats 47/246	easy · avg 71%	▇ 1.0%	at its level
Decision Index 0.4.1: SATA-Bench (case exact accuracy) submitted	20.0% ✎	▁▁▁▁▁▁▁▂▅█▂▁ beats 73/246	hard · avg 22%	▇ 0.9%	typical
Decision Index 0.4.1: ChessBench (accuracy) submitted	33.2% ✎	▄▇▄▄▄▆▄█▄▁▃▁ beats 244/246	hard · avg 16%	▇ 0.5%	blunt
Decision Index 0.4.1: HLE (accuracy) submitted	11.1% ✎	▁▁▂▃▆█▇▅▅▃▁▁ beats 66/246	very hard · avg 13%	▇ 0.0%	blunt

other tests2 not in a task · 7.1% of its score

test	result	vs the others in its run	difficulty	counts	why
Decision Index 0.4.1: private same skills (equated points)	27.3200 points ✎	▂▂▂▃▃▃▄▅▅▇█▃ beats 59/247	avg 41.206 points	▇▇ 4.6%	blunt
Decision Index 0.4.1: private new domains (equated points)	25.7900 points ✎	▃█▃▃▂▆▃▃▄▄▄▄ beats 87/247	avg 38.861 points	▇ 2.5%	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/TheoLeeCJ ↗

$ 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		= 45.5

not counted (2)

Decision Index 0.4.1	Decision Index 0.4.1: RouterBench (selected quality (quality objective)) [correct]	79.5%	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]	56.8%	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