Sieve 4B

+ contribute

Sieve 4B is an open-weights decision model from Sthānika AI, released 13 October 2026. Its published size is 4.66B parameters. Licence: apache-2.0.

Catalogue identity notes

Exact Decision Index tested engine sieve-4b; reported model Sieve 4B. Type: LoRA + head; 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.

53.9±1.2#169 of 355 decision modelsfrom 42 benchmark results across 1 source

$ how it compares

The decision models ranked either side of Sieve 4B, out of 357 with a score, then more from Sthānika AI.

#	model	score	where	released
169	Open-Jev-9B (JevBench mini) Zefan Cai	54±2	source code
170	Solomon v1.1 Doccy Health	54±1	model files	2026-09-21
171	Sieve 4B Sthānika AI	54±1	model files	2026-10-13
172	Bespoke-Nimble-9B (DecisionBench 594dfdcf) Bespoke Labs	54±2	model files
173	JevAny-Qwen-4B (pointer) (JevBench mini) SimpleJev	54±2	source code
	more from Sthānika AI
3	Sieve 31B v3 Sthānika AI	74±2	model files	2026-10-27
37	Sieve 27B v2 Sthānika AI	67±1	model files	2026-10-18
79	Sieve-Vision-27B Sthānika AI	63±4	model files

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

$ what it's good at

tools & automation▁▃▁▁▁▂▃▄▆██▆ #130 of 250 · 7.6% of its score · 6 tests

Beats Tern-35B-A3B, Cerno 4B v1.1, Trio-Spark DI (9B) here, though they rank higher overall. Trails OneJev1.6-0.8B, TokenRhythm NeoHorse-Jev-4B, razorback16 diffgemma (decision readout) 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	73.8% ✎	▁▁▁▁▁▁▁▂▃▃█▂ beats 101/236	easy · avg 70%	▇ 2.3%	sharp · noisy
Decision Index 0.4.1: BFCL (case exact accuracy) submitted	94.4% ✎	▁▂▁▁▁▂▂▄▅▆█▁ beats 120/246	very easy · avg 92%	▇ 2.1%	sharp · maxed out · noisy
Decision Index 0.4.1: Home appliance simulator (case exact accuracy) submitted	26.1% ✎	█▁▁▃▄▄▆▄▁▂▁▃ beats 99/243	medium · avg 37%	▇ 1.1%	sharp · at its level · noisy
Decision Index 0.4.1: When2Call (accuracy) submitted	79.9% ✎	▁▁▁▂▃▅▅▆█▇█▃ beats 138/246	easy · avg 72%	▇ 1.1%	noisy
Decision Index 0.4.1: ToolRet (nDCG@10) submitted	67.3% ✎	▁▁▁▁▁·▁▁▁▄█▁ beats 188/246	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▁▁▁·▁▁▁▁▃▃█▅ #151 of 258 · 6.4% of its score · 7 tests

Beats Quyet-1.0-Large, Decider chat · Qwen3.6-27B (decision readout), Solar Decide here, though they rank higher overall. Trails Amber 0.4B, Saina Helm 2 Omni (0.8B), 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: BRIGHT (nDCG@10) submitted	44.4% ✎	▁▁▁▁▁▁▁▂▂▄▇█ beats 147/246	medium · avg 38%	▇ 2.3%	blunt · at its level
Decision Index 0.4.1: Amazon ESCI (macro-F1) submitted	49.1% ✎	▁▁▁▁▁▁▂▂▃▃█▅ beats 101/246	medium · avg 47%	▇ 2.2%	at its level
Decision Index 0.4.1: HoVer (accuracy) submitted	75.0% ✎	▆▆▇▅▄▄▃▇▄█▆▄ beats 123/246	easy · avg 74%	▇ 0.6%	blunt
Decision Index 0.4.1: PhishNChips phishing decisions (accuracy) submitted	84.3% ✎	█▅▅▄▃▂▁▁▁▁·▁ beats 187/245	easy · avg 70%	▇ 0.5%	at its level · noisy
Decision Index 0.4.1: CLINC150+OOS (macro-F1) submitted	90.2% ✎	▁▁▁▁▁▁▁▂▃▆█▄ beats 130/238	easy · avg 83%	▇ 0.4%	sharp · noisy
Decision Index 0.4.1: BANKING77 (macro-F1) submitted	82.8% ✎	▁·▁▁▁▁▁▁▂█▇█ beats 115/238	easy · avg 79%	▇ 0.3%	noisy
Decision Index 0.4.1: SGD/SGD-X (macro-F1) submitted	38.2% ✎	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▂▂▂▃▂▅▆▃█▆▄▃ #66 of 256 · 23% of its score · 9 tests

Beats Turion Jev 31B, AJev lm6q (Gemma 4 31B), oenogarum (Gemma 4 31B) here, though they rank higher overall.

test	result	vs the others in its run	difficulty	counts	why
Decision Index 0.4.1: HellaSwag (accuracy) submitted	95.0% ✎	▂▄▂▂▄▇▃▇█▅▃▁ beats 195/246	very easy · avg 85%	▇▇ 5.5%	sharp
Decision Index 0.4.1: NLI4CT (macro-F1) submitted	76.8% ✎	▁▁▁▂▂▃▃▄▇▇█▄ beats 113/246	easy · avg 74%	▇▇ 5.4%	typical
Decision Index 0.4.1: ANLI (macro-F1) submitted	63.5% ✎	▁▂▃▅▅█▂····▁ beats 116/246	medium · avg 60%	▇▇ 5.3%	at its level
Decision Index 0.4.1: iSarcasmEval (Sarcasm F1 · track A, English) submitted	46.1% ✎	▁·▁·▁▁▂▄▅█▇▄ beats 114/246	medium · avg 44%	▇ 2.3%	at its level
Decision Index 0.4.1: ACOS (per-review F1) submitted	20.6% ✎	▁··▂▃▃▅▇█▇▃▃ beats 112/246	hard · avg 22%	▇ 1.1%	at its level · noisy
Decision Index 0.4.1: ContractNLI (macro-F1) submitted	79.8% ✎	▁▁▁▁▁▁▁▂▂▇▇█ beats 164/246	easy · avg 71%	▇ 1.1%	at its level · noisy
Decision Index 0.4.1: VAST (macro-F1) submitted	52.4% ✎	▁▁▁▂▅▄▄▂▄▆█▄ beats 86/246	medium · avg 62%	▇ 1.0%	at its level
Decision Index 0.4.1: RAGTruth (F1 on hallucinated class) submitted	77.9% ✎	▁▁▁▁▁▁▁▁▁▃▆█ beats 156/246	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	88.6% ✎	▁···▁▁▁▁▃▆█▃ beats 132/246	easy · avg 84%	▇ 0.7%	noisy

arts & human taste▁▇▄▃▅▅█▅▄▂▁▁ #159 of 247 · 9.2% of its score · 6 tests

Beats Jegue 0.1 (9B), Standard One 8B SH (v4) 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	61.6% ✎	▁▁▂▃▃▃▇▅█▅▃▁ beats 111/246	medium · avg 61%	▇▇ 4.0%	at its level
Decision Index 0.4.1: Humicroedit (accuracy) submitted	59.5% ✎	▁·▁▃▂▂▅█▄▂▁▁ beats 86/246	medium · avg 60%	▇ 1.9%	blunt · at its level
Decision Index 0.4.1: BPoMP (accuracy) submitted	80.4% ✎	▁█▄▄▅█▆▅▄▁▁▁ beats 87/246	easy · avg 83%	▇ 1.2%	sharp · noisy
Decision Index 0.4.1: cfcolor (accuracy) submitted	61.3% ✎	▁········▁▇█ beats 116/246	medium · avg 60%	▇ 1.0%	blunt · at its level
Decision Index 0.4.1: POP909-CL (accuracy) submitted	7.8% ✎	▄▅▅▃▆▅▃▅▂▅▃█ beats 89/237	hard · avg 21%	▇ 0.8%	sharp · at its level · noisy
Decision Index 0.4.1: Habermas Machine (accuracy) submitted	40.0% ✎	▁▂▄█▇▅▁▁▄▄▁▁ beats 62/246	medium · avg 47%	▇ 0.3%	blunt · at its level

knowledge & reasoning▁▂▂▂▂▃▃▅▅▆█▂ #123 of 263 · 45% of its score · 14 tests

Beats OneJev1.6-4B, Winnow-12B (DecisionBench b6ac22b0), Cerno 4B v1.1 here, though they rank higher overall. Trails PurpleMIST-Flash-1.0, JPT-9B Qwen3.5-9B LoRA 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	55.0% ✎	▄▄▃▃▂▄▅▇██▅▁ beats 136/246	medium · avg 46%	▇▇▇▇ 10%	sharp · at its level
Decision Index 0.4.1: MMLU (accuracy) submitted	75.8% ✎	▁▄▄▃▃▂▅▇▅█▇▁ beats 124/246	easy · avg 70%	▇▇▇ 7.1%	sharp
Decision Index 0.4.1: BBH (accuracy) submitted	68.2% ✎	▄▆▄▄█▇▇▄▃▁▁▁ beats 122/246	easy · avg 66%	▇▇▇ 6.4%	at its level
Decision Index 0.4.1: ARC-Challenge (accuracy) submitted	94.4% ✎	▂▂▂▂▂▂▁▃▄▆█▆ beats 132/246	very easy · avg 90%	▇▇ 4.2%	sharp · maxed out
Decision Index 0.4.1: GSM8K (accuracy (4- and 10-choice)) submitted	64.5% ✎	▁··▄█▆▆▆▅▅▃▁ beats 171/246	medium · avg 45%	▇ 3.4%	sharp · at its level
Decision Index 0.4.1: WinoGrande (accuracy) submitted	85.8% ✎	█▅█▅▄█▆▄▃▃▁▁ beats 168/246	easy · avg 78%	▇ 3.2%	typical
Decision Index 0.4.1: GPQA Diamond (accuracy) submitted	42.4% ✎	▂▆▅▅▆█▆▃▁··▁ beats 130/246	medium · avg 39%	▇ 3.1%	blunt · at its level
Decision Index 0.4.1: ARC-Easy (accuracy) submitted	97.9% ✎	▁▁▂▁▂▂▂▂▃▆█▅ beats 122/246	very easy · avg 96%	▇ 2.2%	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	20.6% ✎	▁▁▁▁▁▁▁▂▅█▂▁ beats 79/246	hard · avg 22%	▇ 1.3%	at its level
Decision Index 0.4.1: MuSR (accuracy) submitted	55.0% ✎	▁▁▂█▅▄▁▁▁▁▁▁ beats 69/246	medium · avg 60%	▇ 1.2%	blunt · at its level
Decision Index 0.4.1: CRUXEval (accuracy) submitted	55.8% ✎	██▄▃▃▄▁▁···▁ beats 133/246	medium · avg 58%	▇ 1.1%	at its level · noisy
Decision Index 0.4.1: CLadder (accuracy) submitted	63.9% ✎	▁███▆▅▃▃▁▁▁▁ beats 88/246	easy · avg 71%	▇ 1.0%	at its level
Decision Index 0.4.1: ChessBench (accuracy) submitted	13.8% ✎	▄▇▄▄▄▆▄█▄▁▁▁ beats 99/246	hard · avg 16%	▇ 0.6%	blunt
Decision Index 0.4.1: HLE (accuracy) submitted	9.4% ✎	▁▁▂▃▆█▇▅▅▃▁▁ beats 16/246	very hard · avg 13%	▇ 0.0%	blunt

other tests2 not in a task · 8.1% of its score

test	result	vs the others in its run	difficulty	counts	why
Decision Index 0.4.1: private same skills (equated points)	44.6300 points ✎	▂▂▂▃▃▃▄▅▅▇█▃ beats 122/247	avg 41.206 points	▇▇ 5.2%	blunt
Decision Index 0.4.1: private new domains (equated points)	35.4600 points ✎	▃█▃▃▂▆▃▃▄▄▄▄ beats 105/247	avg 38.861 points	▇ 2.9%	blunt

$ where to run it

Download model files ↗
Published model files verified. Includes adapters or readout weights; follow the model documentation for the required base model. 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/sthanika-ai ↗

$ 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		= 53.9

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]	38.2%	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