Sieve 2B v3

+ contribute

Sieve 2B v3 is an open-weights decision model from Sthānika AI. Its published size is 2.27B parameters. Licence: apache-2.0.

Catalogue identity notes

Exact Decision Index tested engine sieve-2b-v3; reported model Sieve 2B v3. Type: LoRA; 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

· Weights licence ↗ · Parameter count ↗

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

47.5±1.2#227 of 355 decision modelsfrom 42 benchmark results across 1 source

$ how it compares

The decision models ranked either side of Sieve 2B v3, out of 357 with a score, then more from Sthānika AI.

#	model	score	where	released
227	Valen-4B Valen	48±1	model files
228	RSI-Jev v5.0 VL 3B shgao	48±1	model files	2026-10
229	Sieve 2B v3 Sthānika AI	48±1	model files
230	Lev1 3.7B theaviv	47±1	model files	2026-10-04
231	Ryotide Qwen v0.3.1 (decision readout) csabag	47±1	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▁▃▁▁▁▂▃▄▆██▆ #182 of 250 · 8.3% of its score · 6 tests

Beats Decider 4B, JevK5 v0.3, FlyMy.AI Decision 4B v1.6 here, though they rank higher overall. Trails JevK5 Lite, Mola-68M, Mola-150M 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	88.4% ✎	▁▂▁▁▁▂▂▄▅▆█▁ beats 67/246	very easy · avg 92%	▇ 2.9%	sharp · noisy
Decision Index 0.4.1: API-Bank (accuracy) submitted	86.6% ✎	▁▁▁▁▁▁▁▂▃▂█▂ beats 203/236	easy · avg 70%	▇ 2.5%	sharp · at its level · noisy
Decision Index 0.4.1: When2Call (accuracy) submitted	82.0% ✎	▁▁▁▂▃▅▅▆█▇█▃ beats 155/246	easy · avg 72%	▇ 1.1%	at its level · noisy
Decision Index 0.4.1: ToolRet (nDCG@10) submitted	63.0% ✎	▁▁▁▁▁·▁▁▁▄█▁ beats 103/246	medium · avg 60%	▇ 0.9%	blunt · at its level
Decision Index 0.4.1: Home appliance simulator (case exact accuracy) submitted	96.6% ✎	█▁▁▃▄▄▆▄▁▃▁▃ beats 218/243	medium · avg 37%	▇ 0.9%	sharp · at its level · noisy
Decision Index 0.4.1: RouterBench (selected quality (quality objective)) submitted	66.9% ✎	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▁▁▁·▁▁▁▁▃▃█▅ #89 of 258 · 5.8% of its score · 7 tests

Beats AutoTrust-JEV-27B v1.1, Decider 31B (NVFP4), Jev 1.13 here, though they rank higher overall. Trails Jev-Style 2B v4 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	35.3% ✎	▁▁▁▁▁▁▁▂▃▄▇█ beats 73/246	medium · avg 38%	▇ 2.0%	blunt · at its level
Decision Index 0.4.1: Amazon ESCI (macro-F1) submitted	55.7% ✎	▁▁▁▁▁▁▂▂▃▃█▅ beats 164/246	medium · avg 47%	▇ 1.9%	at its level
Decision Index 0.4.1: HoVer (accuracy) submitted	82.9% ✎	▆▆▇▅▄▄▃▇▄█▆▄ beats 170/246	easy · avg 74%	▇ 0.6%	blunt · at its level
Decision Index 0.4.1: PhishNChips phishing decisions (accuracy) submitted	96.4% ✎	█▅▅▄▃▂▁▃▁▁·▁ beats 236/245	easy · avg 70%	▇ 0.5%	at its level · noisy
Decision Index 0.4.1: CLINC150+OOS (macro-F1) submitted	81.8% ✎	▁▁▁▁▁▁▁▂▃▆█▄ beats 72/238	easy · avg 83%	▇ 0.5%	sharp · noisy
Decision Index 0.4.1: BANKING77 (macro-F1) submitted	88.1% ✎	▁·▁▁▁▁▁▁▂█▇█ beats 165/238	easy · avg 79%	▇ 0.3%	noisy
Decision Index 0.4.1: SGD/SGD-X (macro-F1) submitted	43.5% ✎	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▂▂▂▃▂▅▆▃█▆▄▃ #134 of 256 · 24% of its score · 9 tests

Beats Solar Decide, Winnow-12B, JoshuaSP diffgemma (decision readout) 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	87.3% ✎	▂▄▂▂▄▇▃▇█▅▃▁ beats 121/246	very easy · avg 85%	▇▇▇ 7.1%	sharp
Decision Index 0.4.1: NLI4CT (macro-F1) submitted	69.1% ✎	▁▁▁▂▂▃▃▄▇▇█▄ beats 76/246	easy · avg 74%	▇▇ 5.4%	at its level
Decision Index 0.4.1: ANLI (macro-F1) submitted	50.6% ✎	▁▂▃▅▅█▂····▁ beats 71/246	medium · avg 60%	▇▇ 4.9%	at its level
Decision Index 0.4.1: iSarcasmEval (Sarcasm F1 · track A, English) submitted	28.6% ✎	▁·▁·▁▁▂▄▅█▇▄ beats 54/246	medium · avg 44%	▇ 1.9%	at its level
Decision Index 0.4.1: ContractNLI (macro-F1) submitted	80.9% ✎	▁▁▁▁▁▁▁▂▂▇▇█ beats 170/246	easy · avg 71%	▇ 1.1%	at its level · noisy
Decision Index 0.4.1: VAST (macro-F1) submitted	72.6% ✎	▁▁▁▂▅▄▄▂▄▆█▄ beats 151/246	medium · avg 62%	▇ 0.9%	at its level
Decision Index 0.4.1: ACOS (per-review F1) submitted	44.0% ✎	▁··▂▃▃▅▇█▇▃▃ beats 211/246	hard · avg 22%	▇ 0.8%	noisy
Decision Index 0.4.1: RAGTruth (F1 on hallucinated class) submitted	67.9% ✎	▁▁▁▁▁▁▁▁▁▃▆█ beats 88/246	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	84.8% ✎	▁···▁▁▁▁▃▆█▃ beats 97/246	easy · avg 84%	▇ 0.8%	noisy

arts & human taste▁▇▄▃▅▅█▅▄▂▁▁ #81 of 247 · 8.6% of its score · 6 tests

Beats Jev 1.13, jev-brain · Gemma-4-31B (decision readout), Jeffan-4B-2010 here, though they rank higher overall. Trails Amber 0.4B, Crivo-1B, Kas 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	60.0% ✎	▁▁▂▃▃▃▇▅█▅▃▁ beats 101/246	medium · avg 61%	▇▇ 3.8%	at its level
Decision Index 0.4.1: Humicroedit (accuracy) submitted	58.5% ✎	▁·▁▃▂▂▅█▄▂▁▁ beats 81/246	medium · avg 60%	▇ 1.7%	blunt · at its level
Decision Index 0.4.1: BPoMP (accuracy) submitted	91.9% ✎	▁█▄▄▅█▆▅▄▁▁▁ beats 166/246	easy · avg 83%	▇ 1.4%	sharp · noisy
Decision Index 0.4.1: cfcolor (accuracy) submitted	60.2% ✎	▁········▁▇█ beats 104/246	medium · avg 60%	▇ 0.9%	blunt · at its level
Decision Index 0.4.1: POP909-CL (accuracy) submitted	79.3% ✎	▄▅▅▃▆▅▃▅▂▅▃█ beats 215/237	hard · avg 21%	▇ 0.5%	sharp · noisy
Decision Index 0.4.1: Habermas Machine (accuracy) submitted	59.2% ✎	▁▂▄█▇▅▁▁▄▄▁▁ beats 195/246	medium · avg 47%	▇ 0.3%	blunt · at its level

knowledge & reasoning▁▂▂▂▂▃▃▅▅▆█▂ #194 of 263 · 46% of its score · 14 tests

Beats StartLux-Decision v1.1 0.8B here, though they rank higher overall. Trails laya (DecisionBench 1c5edc17), DM-JEPA 1.1, Sieve 2B 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	29.8% ✎	▄▄▃▃▃▄▅▇██▅▁ beats 64/246	medium · avg 46%	▇▇▇▇ 8.8%	sharp · at its level
Decision Index 0.4.1: MMLU (accuracy) submitted	55.9% ✎	▁▄▄▃▃▂▅▇▅█▇▁ beats 66/246	easy · avg 70%	▇▇▇ 7.4%	sharp · at its level
Decision Index 0.4.1: BBH (accuracy) submitted	60.8% ✎	▄▆▄▄█▇▇▄▃▁▁▁ beats 84/246	easy · avg 66%	▇▇▇ 6.3%	at its level
Decision Index 0.4.1: ARC-Challenge (accuracy) submitted	78.8% ✎	▂▂▂▂▂▃▁▃▄▆█▆ beats 69/246	very easy · avg 90%	▇▇ 6.2%	sharp
Decision Index 0.4.1: ARC-Easy (accuracy) submitted	91.4% ✎	▁▁▂▁▂▂▃▂▃▆█▅ beats 71/246	very easy · avg 96%	▇▇ 3.8%	sharp · maxed out · noisy
Decision Index 0.4.1: WinoGrande (accuracy) submitted	77.5% ✎	█▅█▅▄█▆▄▃▃▁▁ beats 117/246	easy · avg 78%	▇ 3.4%	typical
Decision Index 0.4.1: GSM8K (accuracy (4- and 10-choice)) submitted	39.0% ✎	▁··▄█▆▆▆▅▅▃▁ beats 105/246	medium · avg 45%	▇ 2.9%	sharp · at its level
Decision Index 0.4.1: GPQA Diamond (accuracy) submitted	27.0% ✎	▂▆▅▅▆█▆▃▁··▁ beats 35/246	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	74.2% ✎	▁▁▂█▅▄▃▁▁▁▁▁ beats 230/246	medium · avg 60%	▇ 1.1%	blunt · at its level
Decision Index 0.4.1: CRUXEval (accuracy) submitted	48.4% ✎	██▄▃▃▄▁▁···▁ beats 102/246	medium · avg 58%	▇ 1.0%	at its level · noisy
Decision Index 0.4.1: SATA-Bench (case exact accuracy) submitted	28.7% ✎	▁▁▁▁▁▁▁▂▅█▂▁ beats 129/246	hard · avg 22%	▇ 1.0%	typical
Decision Index 0.4.1: CLadder (accuracy) submitted	88.4% ✎	▁███▆▅▃▃▁▁▁▁ beats 215/246	easy · avg 71%	▇ 0.9%	at its level
Decision Index 0.4.1: ChessBench (accuracy) submitted	22.5% ✎	▄▇▄▄▄▆▄█▄▁▁▁ beats 195/246	hard · avg 16%	▇ 0.5%	blunt
Decision Index 0.4.1: HLE (accuracy) submitted	12.8% ✎	▁▁▂▃▆█▇▅▅▃▁▁ beats 136/246	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)	45.2700 points ✎	▂▂▂▃▃▃▄▅▅▇█▃ beats 123/247	avg 41.206 points	▇▇ 4.6%	blunt
Decision Index 0.4.1: private new domains (equated points)	18.2200 points ✎	▃█▃▃▂▆▃▃▄▄▄▄ beats 69/247	avg 38.861 points	▇ 2.5%	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		= 47.5

not counted (2)

Decision Index 0.4.1	Decision Index 0.4.1: RouterBench (selected quality (quality objective)) [correct]	66.9%	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]	43.5%	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