Laya

+ contribute

Laya is an open-weights decision model from ConvAI Innovations. Its published size is 0.42B parameters. Licence: apache-2.0.

Catalogue identity notes

Exact Decision Index tested engine laya-1e28ac20; reported model Laya. Type: full fine-tune; technique: encoder. Implementation: https://github.com/NandhaKishorM/laya. 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.

21.2±1.1#338 of 355 decision modelsfrom 42 benchmark results across 1 source

$ how it compares

The decision models ranked either side of Laya, out of 357 with a score, then more from ConvAI Innovations.

#	model	score	where	RAM
338	gliner2.5-small-v1 (DecisionBench 7e6f537f) Fastino	23±2	model files
339	Qwen-2.5-1B-RLCD (decision readout) harshatheg	21±1	source code
340	Laya ConvAI Innovations	21±1	model files
341	Verdict heman10x	21±1	model files
342	jeff GLiFormer-large (decision readout) logan-markewich	21±1	source code
	more from ConvAI Innovations
315	laya-typed-decisions (DecisionBench bc76315b) ConvAI Innovations	29±2	model files
331	laya (DecisionBench 1c5edc17) ConvAI Innovations	25±1	local	~1.7GB

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

$ what it's good at

tools & automation▁▃▁▁▁▂▃▄▆██▆ #241 of 250 · 10% of its score · 6 tests

Beats DM-JEPA 1.1, Von 1.3, Bekko System One v0 68M here, though they rank higher overall. Trails LFM2.5-350M-RLCD, CLM-v0.1-8B Qwen3-8B heads, Verdict 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	38.1% ✎	▁▃▁▁▁▂▂▄▅▆█▁ beats 9/246	very easy · avg 92%	▇▇▇ 6.9%	sharp · at its level · noisy
Decision Index 0.4.1: API-Bank (accuracy) submitted	11.5% ✎	▁▁▁▁▃▁▁▂▃▂█▂ beats 15/236	easy · avg 70%	▇ 1.6%	sharp · noisy
Decision Index 0.4.1: When2Call (accuracy) submitted	11.9% ✎	▃▁▁▂▃▅▅▆█▇█▃ beats 0/246	easy · avg 72%	▇ 1.0%	at its level · noisy
Decision Index 0.4.1: ToolRet (nDCG@10) submitted	12.8% ✎	▁▃▁▁▁·▁▁▁▄█▁ beats 4/246	medium · avg 60%	▇ 0.7%	blunt · at its level
Decision Index 0.4.1: Home appliance simulator (case exact accuracy) submitted	0.0% ✎	█▁▁▃▄▄▆▄▁▂▁▃ beats 26/243	medium · avg 37%	▇ 0.1%	sharp · too hard for it · noisy
Decision Index 0.4.1: RouterBench (selected quality (quality objective)) submitted	57.1% ✎	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▁▁▁·▁▃▁▁▃▃█▅ #242 of 258 · 4.5% of its score · 7 tests

Beats DM-JEPA 1.1, Together Tev1 Qwen3.5-0.8B, InternLM Intern-Decision Qwen3.5-0.8B here, though they rank higher overall.

test	result	vs the others in its run	difficulty	counts	why
Decision Index 0.4.1: Amazon ESCI (macro-F1) submitted	24.4% ✎	▁▁▁▁▁▁▃▂▃▃█▅ beats 17/246	medium · avg 47%	▇ 1.2%	at its level
Decision Index 0.4.1: BRIGHT (nDCG@10) submitted	19.9% ✎	▁▁▁▁▁▃▁▂▂▄▇█ beats 17/246	medium · avg 38%	▇ 1.2%	blunt
Decision Index 0.4.1: CLINC150+OOS (macro-F1) submitted	3.2% ✎	▁▁▃▁▁▁▁▂▃▆█▄ beats 5/238	easy · avg 83%	▇ 0.6%	sharp · at its level · noisy
Decision Index 0.4.1: HoVer (accuracy) submitted	55.8% ✎	▆▆▇▅▄▄▃▇▄█▆▄ beats 26/246	easy · avg 74%	▇ 0.6%	blunt · at its level
Decision Index 0.4.1: PhishNChips phishing decisions (accuracy) submitted	50.1% ✎	█▅▅▄▃▂▁▁▁▁·▁ beats 29/245	easy · avg 70%	▇ 0.5%	at its level · noisy
Decision Index 0.4.1: BANKING77 (macro-F1) submitted	14.3% ✎	▁·▁▁▁▃▁▁▂█▇█ beats 7/238	easy · avg 79%	▇ 0.4%	at its level · noisy
Decision Index 0.4.1: SGD/SGD-X (macro-F1) submitted	42.4% ✎	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▃▂▂▃▂▅▆▃█▆▄▃ #246 of 256 · 23% of its score · 9 tests

Beats Bosun v3.1 0.6B LoRA, Crivo-17M, GLiNER2.5-Decide here, though they rank higher overall. Trails CLM-v0.1-8B Qwen3-8B heads, jeff GLiFormer-large (decision readout), Verdict 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	33.1% ✎	▃▄▂▂▄▇▃▇█▅▃▁ beats 11/246	very easy · avg 85%	▇▇▇▇ 9.2%	sharp · at its level
Decision Index 0.4.1: NLI4CT (macro-F1) submitted	47.7% ✎	▁▁▁▃▂▃▃▄▇▇█▄ beats 9/246	easy · avg 74%	▇▇ 5.6%	at its level
Decision Index 0.4.1: ANLI (macro-F1) submitted	48.7% ✎	▁▂▃▅▅█▂····▁ beats 61/246	medium · avg 60%	▇ 3.5%	at its level
Decision Index 0.4.1: FinEntity (macro-F1) submitted	61.0% ✎	▁···▁▁▁▃▃▆█▃ beats 24/246	easy · avg 84%	▇ 1.1%	at its level · noisy
Decision Index 0.4.1: ContractNLI (macro-F1) submitted	29.0% ✎	▁▁▁▁▃▁▁▂▂▇▇█ beats 14/246	easy · avg 71%	▇ 1.0%	at its level · noisy
Decision Index 0.4.1: iSarcasmEval (Sarcasm F1 · track A, English) submitted	14.9% ✎	▁·▁·▁▁▃▄▅█▇▄ beats 13/246	medium · avg 44%	▇ 0.9%	typical
Decision Index 0.4.1: VAST (macro-F1) submitted	40.5% ✎	▁▁▁▂▅▄▄▂▄▆█▄ beats 33/246	medium · avg 62%	▇ 0.7%	at its level
Decision Index 0.4.1: RAGTruth (F1 on hallucinated class) submitted	48.8% ✎	▁▁▁▁▁▁▁▁▃▃▆█ beats 43/246	easy · avg 66%	▇ 0.6%	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	3.5% ✎	▁··▃▃▃▅▇█▇▃▃ beats 13/246	hard · avg 22%	▇ 0.2%	too hard for it · noisy

arts & human taste▁▇▄▃▅▅█▅▄▂▁▁ #220 of 247 · 7.1% of its score · 6 tests

Beats Valen-4B, TokenRhythm NeoHorse-Jev-4B, OneJev1.6-0.8B here, though they rank higher overall. Trails LFM2.5-350M-RLCD, system-one-gemma, Bekko System One v0 17M 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	27.1% ✎	▁▃▂▃▃▃▇▅█▅▃▁ beats 12/246	medium · avg 61%	▇ 2.7%	at its level
Decision Index 0.4.1: BPoMP (accuracy) submitted	51.6% ✎	▁█▄▄▅█▆▅▄▁▁▁ beats 28/246	easy · avg 83%	▇ 1.9%	sharp · at its level · noisy
Decision Index 0.4.1: Humicroedit (accuracy) submitted	47.2% ✎	▁·▃▃▂▂▅█▄▂▁▁ beats 6/246	medium · avg 60%	▇ 1.4%	blunt · at its level
Decision Index 0.4.1: cfcolor (accuracy) submitted	52.3% ✎	▁········▁▇█ beats 39/246	medium · avg 60%	▇ 0.8%	blunt · at its level
Decision Index 0.4.1: Habermas Machine (accuracy) submitted	33.4% ✎	▁▃▄█▇▅▁▁▄▄▁▁ beats 11/246	medium · avg 47%	▇ 0.2%	blunt · at its level
Decision Index 0.4.1: POP909-CL (accuracy) submitted	5.1% ✎	▄▅▅▃▆▅▃▅▂▅▃█ beats 73/237	hard · avg 21%	▇ 0.1%	sharp · too hard for it · noisy

knowledge & reasoning▁▃▂▂▂▃▃▅▅▆█▂ #255 of 263 · 49% of its score · 14 tests

Beats Crivo-17M, Prumo (rule-based engine) here, though they rank higher overall. Trails CLM-v0.1-8B Qwen3-8B heads, jeff GLiFormer-large (decision readout), Verdict here, though they rank lower overall.

test	result	vs the others in its run	difficulty	counts	why
Decision Index 0.4.1: ARC-Easy (accuracy) submitted	47.0% ✎	▁▃▂▁▂▂▂▂▃▆█▅ beats 8/246	very easy · avg 96%	▇▇▇▇▇▇ 15%	sharp · at its level · noisy
Decision Index 0.4.1: ARC-Challenge (accuracy) submitted	28.6% ✎	▃▂▂▂▂▂▁▃▄▆█▆ beats 7/246	very easy · avg 90%	▇▇▇▇ 10%	sharp · at its level
Decision Index 0.4.1: MMLU (accuracy) submitted	30.7% ✎	▁▄▄▃▃▂▅▇▅█▇▁ beats 9/246	easy · avg 70%	▇▇ 5.8%	sharp · at its level
Decision Index 0.4.1: BBH (accuracy) submitted	34.1% ✎	▄▆▄▄█▇▇▄▃▁▁▁ beats 13/246	easy · avg 66%	▇▇ 4.7%	at its level
Decision Index 0.4.1: WinoGrande (accuracy) submitted	50.5% ✎	█▅█▅▄█▆▄▃▃▁▁ beats 10/246	easy · avg 78%	▇▇ 3.8%	at its level
Decision Index 0.4.1: MMLU-Pro (accuracy) submitted	13.6% ✎	▄▄▃▃▂▄▅▇██▅▁ beats 13/246	medium · avg 46%	▇ 3.4%	sharp
Decision Index 0.4.1: GPQA Diamond (accuracy) submitted	27.6% ✎	▂▆▅▅▆█▆▃▁··▁ beats 42/246	medium · avg 39%	▇ 1.6%	blunt
Decision Index 0.4.1: GSM8K (accuracy (4- and 10-choice)) submitted	16.9% ✎	▁··▄█▆▆▆▅▅▃▁ beats 3/246	medium · avg 45%	▇ 1.0%	sharp

6 more tests

test	result	vs the others in its run	difficulty	counts	why
Decision Index 0.4.1: MuSR (accuracy) submitted	43.2% ✎	▁▃▂█▅▄▁▁▁▁▁▁ beats 17/246	medium · avg 60%	▇ 0.9%	blunt · at its level
Decision Index 0.4.1: CLadder (accuracy) submitted	52.9% ✎	▁███▆▅▃▃▁▁▁▁ beats 27/246	easy · avg 71%	▇ 0.9%	at its level
Decision Index 0.4.1: CRUXEval (accuracy) submitted	40.2% ✎	██▄▃▃▄▁▁···▁ beats 57/246	medium · avg 58%	▇ 0.7%	at its level · noisy
Decision Index 0.4.1: SATA-Bench (case exact accuracy) submitted	0.3% ✎	▃▁▁▁▁▁▁▂▅█▂▁ beats 10/246	hard · avg 22%	▇ 0.3%	too hard for it
Decision Index 0.4.1: ChessBench (accuracy) submitted	7.7% ✎	▄▇▄▄▄▆▄█▄▁▁▁ beats 16/246	hard · avg 16%	▇ 0.2%	blunt
Decision Index 0.4.1: HLE (accuracy) submitted	14.0% ✎	▁▁▂▃▆█▇▅▅▃▁▁ beats 177/246	very hard · avg 13%	▇ 0.0%	blunt

other tests2 not in a task · 5.9% of its score

test	result	vs the others in its run	difficulty	counts	why
Decision Index 0.4.1: private same skills (equated points)	8.1300 points ✎	▂▃▂▃▃▃▄▅▅▇█▃ beats 12/247	avg 41.206 points	▇▇ 3.8%	blunt
Decision Index 0.4.1: private new domains (equated points)	10.8400 points ✎	▃█▃▃▂▆▃▃▄▄▄▄ beats 24/247	avg 38.861 points	▇ 2.1%	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/convaiinnovations ↗

$ how the score was worked out · Decision Index 0.4.1 100%

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	▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇ 100%	CC BY 4.0
prior · pull toward the middle		▇ 0%
weighted average		= 21.2

not counted (2)

Decision Index 0.4.1	Decision Index 0.4.1: RouterBench (selected quality (quality objective)) [correct]	57.1%	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]	42.4%	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