Sieve 31B v3

+ contribute

Sieve 31B v3 is an open-weights decision model from Sthānika AI, released 27 October 2026. Its published size is 31.27B parameters. Licence: apache-2.0.

Catalogue identity notes

Exact Decision Index tested engine sieve-31b-v3; reported model Sieve 31B v3. Type: full fine-tune; 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.

73.7±1.6#3 of 355 decision modelsfrom 42 benchmark results across 1 source

$ how it compares

The decision models ranked either side of Sieve 31B v3, out of 357 with a score, then more from Sthānika AI.

#	model	score	where	released
1	Jeffan-31b-1710 tattrongvu	76±2	model files
2	baselabs mammoth-1 (27B) baselabs	75±1	model files	2026-10-21
3	Sieve 31B v3 Sthānika AI	74±2	model files	2026-10-27
4	Fastino GLiDE 1.2 Fastino	73±1	model files	2026-10-22
5	KnowLine 31B Gen1 PelaAI	73±2	model files	2026-10-24
	more from Sthānika AI
37	Sieve 27B v2 Sthānika AI	67±1	model files	2026-10-18
79	Sieve-Vision-27B Sthānika AI	63±4	model files
137	Sieve 9B v2 Sthānika AI	57±1	model files	2026-10-13

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

$ what it's good at

tools & automation▁▃▁▁▁▂▃▄▆██▆ #21 of 250 · 4.8% of its score · 6 tests

Beats Jeffan-31b-1710 here, though they rank higher overall. Trails Solomon v1.1, Clef-flash, Clef-omni 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	88.8% ✎	▁▁▁▁▁▁▁▂▃▂█▃ beats 223/236	easy · avg 70%	▇ 1.4%	sharp · maxed out · noisy
Decision Index 0.4.1: ToolRet (nDCG@10) submitted	70.5% ✎	▁▁▁▁▁·▁▁▁▄█▃ beats 235/246	medium · avg 60%	▇ 1.2%	blunt · at its level
Decision Index 0.4.1: When2Call (accuracy) submitted	88.0% ✎	▁▁▁▂▃▅▅▆█▇█▃ beats 196/246	easy · avg 72%	▇ 0.9%	noisy
Decision Index 0.4.1: Home appliance simulator (case exact accuracy) submitted	96.6% ✎	█▁▁▃▄▄▆▄▁▃▁▃ beats 217/243	medium · avg 37%	▇ 0.7%	sharp · noisy
Decision Index 0.4.1: BFCL (case exact accuracy) submitted	97.9% ✎	▁▂▁▁▁▂▂▄▅▆█▁ beats 223/246	very easy · avg 92%	▇ 0.6%	sharp · maxed out · noisy
Decision Index 0.4.1: RouterBench (selected quality (quality objective)) submitted	79.8% ✎	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▁▁▁·▁▁▁▁▃▃█▅ #2 of 258 · 8.6% of its score · 7 tests

Beats Jeffan-31b-1710, baselabs mammoth-1 (27B) here, though they rank higher overall. Trails StartLux-Decision v1.1 27B 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	50.6% ✎	▁▁▁▁▁▁▁▂▂▄▇█ beats 241/246	medium · avg 38%	▇▇ 3.8%	blunt · at its level
Decision Index 0.4.1: Amazon ESCI (macro-F1) submitted	64.5% ✎	▁▁▁▁▁▁▂▂▃▃█▅ beats 244/246	medium · avg 47%	▇ 3.1%	at its level
Decision Index 0.4.1: HoVer (accuracy) submitted	89.1% ✎	▆▆▇▅▄▄▃▇▄█▆▄ beats 233/246	easy · avg 74%	▇ 0.6%	blunt
Decision Index 0.4.1: PhishNChips phishing decisions (accuracy) submitted	94.9% ✎	█▅▅▄▃▂▃▁▁▁·▁ beats 229/245	easy · avg 70%	▇ 0.6%	noisy
Decision Index 0.4.1: BANKING77 (macro-F1) submitted	91.3% ✎	▁·▁▁▁▁▁▁▂█▇█ beats 203/238	easy · avg 79%	▇ 0.3%	noisy
Decision Index 0.4.1: CLINC150+OOS (macro-F1) submitted	96.8% ✎	▁▁▁▁▁▁▁▂▃▆█▄ beats 231/238	easy · avg 83%	▇ 0.2%	sharp · maxed out · noisy
Decision Index 0.4.1: SGD/SGD-X (macro-F1) submitted	36.6% ✎	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▂▂▂▃▂▅▆▃█▆▄▃ #15 of 256 · 21% of its score · 9 tests

Trails Clef-omni, Clef, Torchcast Decision 27B here, though they rank lower overall.

test	result	vs the others in its run	difficulty	counts	why
Decision Index 0.4.1: ANLI (macro-F1) submitted	80.9% ✎	▁▂▃▅▅█▃····▁ beats 243/246	medium · avg 60%	▇▇ 5.4%	typical
Decision Index 0.4.1: NLI4CT (macro-F1) submitted	84.2% ✎	▁▁▁▂▂▃▃▄▇▇█▄ beats 204/246	easy · avg 74%	▇▇ 5.0%	typical
Decision Index 0.4.1: iSarcasmEval (Sarcasm F1 · track A, English) submitted	68.3% ✎	▁·▁·▁▁▂▄▅█▇▄ beats 221/246	medium · avg 44%	▇ 3.0%	at its level
Decision Index 0.4.1: ACOS (per-review F1) submitted	50.1% ✎	▁··▂▃▃▅▇█▇▃▃ beats 223/246	hard · avg 22%	▇ 2.3%	at its level · noisy
Decision Index 0.4.1: HellaSwag (accuracy) submitted	97.5% ✎	▂▄▂▂▄▇▃▇█▅▃▁ beats 230/246	very easy · avg 85%	▇ 2.0%	sharp · maxed out
Decision Index 0.4.1: VAST (macro-F1) submitted	79.7% ✎	▁▁▁▂▅▄▄▂▄▆█▄ beats 220/246	medium · avg 62%	▇ 1.1%	typical
Decision Index 0.4.1: ContractNLI (macro-F1) submitted	79.8% ✎	▁▁▁▁▁▁▁▂▂▇▇█ beats 162/246	easy · avg 71%	▇ 1.0%	noisy
Decision Index 0.4.1: RAGTruth (F1 on hallucinated class) submitted	82.1% ✎	▁▁▁▁▁▁▁▁▁▃▆█ beats 202/246	easy · avg 66%	▇ 0.7%	noisy

1 more test

test	result	vs the others in its run	difficulty	counts	why
Decision Index 0.4.1: FinEntity (macro-F1) submitted	94.2% ✎	▁···▁▁▁▁▃▆█▃ beats 215/246	easy · avg 84%	▇ 0.4%	maxed out · noisy

arts & human taste▁▇▄▃▅▅█▅▄▂▁▃ #2 of 247 · 11% of its score · 6 tests

Beats Jeffan-31b-1710, baselabs mammoth-1 (27B) here, though they rank higher overall. Trails vera-large 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	87.3% ✎	▁▁▂▃▃▃▇▅█▅▃▃ beats 243/246	medium · avg 61%	▇▇ 3.9%	typical
Decision Index 0.4.1: Humicroedit (accuracy) submitted	67.6% ✎	▁·▁▃▂▂▅█▄▃▁▁ beats 228/246	medium · avg 60%	▇ 2.7%	blunt · at its level
Decision Index 0.4.1: cfcolor (accuracy) submitted	66.5% ✎	▁········▁▇█ beats 217/246	medium · avg 60%	▇ 1.5%	blunt · at its level
Decision Index 0.4.1: POP909-CL (accuracy) submitted	79.3% ✎	▄▅▅▃▆▅▃▅▂▅▃█ beats 214/237	hard · avg 21%	▇ 1.3%	sharp · at its level · noisy
Decision Index 0.4.1: BPoMP (accuracy) submitted	99.2% ✎	▁█▄▄▅█▆▅▄▁▁▃ beats 245/246	easy · avg 83%	▇ 0.7%	sharp · maxed out · noisy
Decision Index 0.4.1: Habermas Machine (accuracy) submitted	62.2% ✎	▁▂▄█▇▅▁▁▄▄▁▁ beats 220/246	medium · avg 47%	▇ 0.5%	blunt · at its level

knowledge & reasoning▁▂▂▂▂▃▃▅▅▆█▃ #5 of 263 · 41% of its score · 14 tests

Trails Jev 1.13, Perplexity Decider v1.1 (27B) 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	75.8% ✎	▄▄▃▃▂▄▅▇██▅▁ beats 240/246	medium · avg 46%	▇▇▇▇ 11%	sharp
Decision Index 0.4.1: BBH (accuracy) submitted	89.6% ✎	▄▆▄▄█▇▇▄▃▁▁▁ beats 233/246	easy · avg 66%	▇▇ 5.4%	typical
Decision Index 0.4.1: MMLU (accuracy) submitted	89.7% ✎	▁▄▄▃▃▂▅▇▅█▇▁ beats 239/246	easy · avg 70%	▇▇ 5.0%	sharp
Decision Index 0.4.1: GPQA Diamond (accuracy) submitted	54.6% ✎	▂▆▅▅▆█▆▃▁··▁ beats 228/246	medium · avg 39%	▇▇ 4.9%	blunt · at its level
Decision Index 0.4.1: GSM8K (accuracy (4- and 10-choice)) submitted	91.0% ✎	▁··▄█▆▆▆▅▅▃▃ beats 243/246	medium · avg 45%	▇ 3.5%	sharp
Decision Index 0.4.1: SATA-Bench (case exact accuracy) submitted	39.4% ✎	▁▁▁▁▁▁▁▂▅█▃▁ beats 223/246	hard · avg 22%	▇ 2.7%	at its level
Decision Index 0.4.1: WinoGrande (accuracy) submitted	94.9% ✎	█▅█▅▄█▆▄▃▃▁▁ beats 231/246	easy · avg 78%	▇ 2.4%	typical
Decision Index 0.4.1: MuSR (accuracy) submitted	80.8% ✎	▁▁▂█▅▄▁▁▃▁▁▁ beats 234/246	medium · avg 60%	▇ 1.6%	blunt · at its level

6 more tests

test	result	vs the others in its run	difficulty	counts	why
Decision Index 0.4.1: ChessBench (accuracy) submitted	25.5% ✎	▄▇▄▄▄▆▄█▄▁▁▁ beats 226/246	hard · avg 16%	▇ 1.3%	blunt · at its level
Decision Index 0.4.1: CRUXEval (accuracy) submitted	93.2% ✎	██▄▃▃▄▃▁···▁ beats 243/246	medium · avg 58%	▇ 1.2%	noisy
Decision Index 0.4.1: ARC-Challenge (accuracy) submitted	97.8% ✎	▂▂▂▂▂▂▁▃▄▆█▆ beats 235/246	very easy · avg 90%	▇ 1.1%	sharp · maxed out
Decision Index 0.4.1: CLadder (accuracy) submitted	92.3% ✎	▁███▆▅▃▃▃▁▁▁ beats 227/246	easy · avg 71%	▇ 1.0%	typical
Decision Index 0.4.1: ARC-Easy (accuracy) submitted	99.3% ✎	▁▁▂▁▂▂▂▂▃▆█▅ beats 236/246	very easy · avg 96%	▇ 0.4%	sharp · maxed out · noisy
Decision Index 0.4.1: HLE (accuracy) submitted	13.0% ✎	▁▁▂▃▆█▇▅▅▃▁▁ beats 146/246	very hard · avg 13%	▇ 0.0%	blunt

other tests2 not in a task · 13% of its score

test	result	vs the others in its run	difficulty	counts	why
Decision Index 0.4.1: private same skills (equated points)	68.2700 points ✎	▂▂▂▃▃▃▄▅▅▇█▃ beats 246/247	avg 41.206 points	▇▇▇ 8.2%	blunt
Decision Index 0.4.1: private new domains (equated points)	75.3300 points ✎	▃█▃▃▂▆▃▃▄▄▄▄ beats 240/247	avg 38.861 points	▇▇ 4.5%	blunt

$ where to run it

Download model files ↗
Published model files verified. 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		= 73.7

not counted (2)

Decision Index 0.4.1	Decision Index 0.4.1: RouterBench (selected quality (quality objective)) [correct]	79.8%	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]	36.6%	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