crystal-0.8b

+ contribute

crystal-0.8b is an open-weights decision model from Kyunghwan-kim. Its published size is 0.85B parameters. Licence: cc-by-nc-4.0.

Catalogue identity notes

Exact Decision Index tested engine crystal-0.8b; reported model crystal-0.8b. 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

· Weights licence ↗ · Parameter count ↗

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

44.3±1.1#247 of 355 decision modelsfrom 42 benchmark results across 1 source

$ how it compares

The decision models ranked either side of crystal-0.8b, out of 357 with a score.

#	model	score	where	RAM
247	GLiClass v4 edge (0.4B) Knowledgator Engineering	45±1	local	~2.7GB
248	Jevstral 8B AmirBraham	44±1	model files
249	crystal-0.8b Kyunghwan-kim	44±1	model files
250	QEV-0.8B ken-jo	44±4	model files
251	openvons Qwen3-4B (decision readout) genai-craft	43±1	source code

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

$ what it's good at

tools & automation▁▃▁▁▁▂▃▄▆██▆ #177 of 250 · 8.6% of its score · 6 tests

Beats LiquidAI d1-3B, Decider 4B, JevK5 v0.3 here, though they rank higher overall. Trails JevK5 Lite, Mola-150M, Gevva E2B 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	89.6% ✎	▁▂▁▁▁▂▂▄▅▆█▁ beats 71/246	very easy · avg 92%	▇ 3.5%	sharp · noisy
Decision Index 0.4.1: API-Bank (accuracy) submitted	82.1% ✎	▁▁▁▁▁▁▁▂▃▂█▂ beats 127/236	easy · avg 70%	▇ 2.5%	sharp · at its level · noisy
Decision Index 0.4.1: When2Call (accuracy) submitted	82.3% ✎	▁▁▁▂▃▅▅▆█▇█▃ beats 157/246	easy · avg 72%	▇ 1.1%	at its level · noisy
Decision Index 0.4.1: ToolRet (nDCG@10) submitted	63.6% ✎	▁▁▁▁▁·▁▁▁▄█▁ beats 112/246	medium · avg 60%	▇ 0.8%	blunt · at its level
Decision Index 0.4.1: Home appliance simulator (case exact accuracy) submitted	14.8% ✎	█▁▁▃▄▄▆▄▁▂▁▃ beats 77/243	medium · avg 37%	▇ 0.7%	sharp · noisy
Decision Index 0.4.1: RouterBench (selected quality (quality objective)) submitted	71.2% ✎	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▁▁▁·▁▁▁▁▃▃█▅ #104 of 258 · 5.5% of its score · 7 tests

Beats AutoTrust-JEV-27B v1.1, Decider 31B (NVFP4), Quyet-1.0-Large here, though they rank higher overall.

test	result	vs the others in its run	difficulty	counts	why
Decision Index 0.4.1: BRIGHT (nDCG@10) submitted	37.3% ✎	▁▁▁▁▁▁▁▂▂▄▇█ beats 82/246	medium · avg 38%	▇ 1.8%	blunt · at its level
Decision Index 0.4.1: Amazon ESCI (macro-F1) submitted	54.9% ✎	▁▁▁▁▁▁▂▂▃▃█▅ beats 150/246	medium · avg 47%	▇ 1.8%	at its level
Decision Index 0.4.1: HoVer (accuracy) submitted	84.8% ✎	▆▆▇▅▄▄▃▇▄█▆▄ beats 180/246	easy · avg 74%	▇ 0.6%	blunt · at its level
Decision Index 0.4.1: PhishNChips phishing decisions (accuracy) submitted	51.9% ✎	█▅▅▄▃▂▁▁▁▁·▁ beats 56/245	easy · avg 70%	▇ 0.5%	at its level · noisy
Decision Index 0.4.1: CLINC150+OOS (macro-F1) submitted	89.0% ✎	▁▁▁▁▁▁▁▂▃▆█▄ beats 110/238	easy · avg 83%	▇ 0.5%	sharp · noisy
Decision Index 0.4.1: BANKING77 (macro-F1) submitted	89.4% ✎	▁·▁▁▁▁▁▁▂█▇█ beats 171/238	easy · avg 79%	▇ 0.3%	noisy
Decision Index 0.4.1: SGD/SGD-X (macro-F1) submitted	38.9% ✎	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▂▂▂▃▂▅▆▃█▆▄▃ #161 of 256 · 24% 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	82.3% ✎	▂▄▂▂▄▇▃▇█▅▃▁ beats 95/246	very easy · avg 85%	▇▇▇ 7.9%	sharp
Decision Index 0.4.1: NLI4CT (macro-F1) submitted	68.4% ✎	▁▁▁▂▂▃▃▄▇▇█▄ beats 69/246	easy · avg 74%	▇▇ 5.4%	at its level
Decision Index 0.4.1: ANLI (macro-F1) submitted	50.1% ✎	▁▂▃▅▅█▂····▁ beats 70/246	medium · avg 60%	▇▇ 4.7%	at its level
Decision Index 0.4.1: iSarcasmEval (Sarcasm F1 · track A, English) submitted	23.8% ✎	▁·▁·▁▁▂▄▅█▇▄ beats 31/246	medium · avg 44%	▇ 1.8%	at its level
Decision Index 0.4.1: ContractNLI (macro-F1) submitted	83.8% ✎	▁▁▁▁▁▁▁▂▂▇▇█ beats 203/246	easy · avg 71%	▇ 1.0%	at its level · noisy
Decision Index 0.4.1: VAST (macro-F1) submitted	42.6% ✎	▁▁▁▂▅▄▄▂▄▆█▄ beats 43/246	medium · avg 62%	▇ 0.9%	at its level
Decision Index 0.4.1: FinEntity (macro-F1) submitted	90.3% ✎	▁···▁▁▁▁▃▆█▃ beats 150/246	easy · avg 84%	▇ 0.8%	noisy
Decision Index 0.4.1: RAGTruth (F1 on hallucinated class) submitted	77.2% ✎	▁▁▁▁▁▁▁▁▁▃▆█ beats 147/246	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	31.4% ✎	▁··▂▃▃▅▇█▇▃▃ beats 169/246	hard · avg 22%	▇ 0.7%	noisy

arts & human taste▁▇▄▃▅▅█▅▄▂▁▁ #142 of 247 · 8.3% of its score · 6 tests

Beats OneJev1.6-4B, Nace.AI Drex v1.5, Jegue 0.1 (9B) here, though they rank higher overall. Trails Crivo-32M v2, Crivo-68M, Crivo-150M v2 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	61.2% ✎	▁▁▂▃▃▃▇▅█▅▃▁ beats 107/246	medium · avg 61%	▇ 3.6%	at its level
Decision Index 0.4.1: Humicroedit (accuracy) submitted	62.2% ✎	▁·▁▃▂▂▅█▄▂▁▁ beats 155/246	medium · avg 60%	▇ 1.6%	blunt · at its level
Decision Index 0.4.1: BPoMP (accuracy) submitted	83.9% ✎	▁█▄▄▅█▆▅▄▁▁▁ beats 105/246	easy · avg 83%	▇ 1.5%	sharp · noisy
Decision Index 0.4.1: cfcolor (accuracy) submitted	58.5% ✎	▁········▁▇█ beats 86/246	medium · avg 60%	▇ 0.9%	blunt · at its level
Decision Index 0.4.1: POP909-CL (accuracy) submitted	1.6% ✎	▄▅▅▃▆▅▃▅▂▅▃█ beats 36/237	hard · avg 21%	▇ 0.4%	sharp · noisy
Decision Index 0.4.1: Habermas Machine (accuracy) submitted	58.0% ✎	▁▂▄█▇▅▁▁▄▄▁▁ beats 190/246	medium · avg 47%	▇ 0.3%	blunt · at its level

knowledge & reasoning▁▂▂▂▂▃▃▅▅▆█▂ #196 of 263 · 46% of its score · 14 tests

Beats OneJev1.6-0.8B, GLiClass v4 edge (0.4B) 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	34.1% ✎	▄▄▃▃▃▄▅▇██▅▁ beats 72/246	medium · avg 46%	▇▇▇ 8.0%	sharp · at its level
Decision Index 0.4.1: MMLU (accuracy) submitted	55.4% ✎	▁▄▄▃▃▂▅▇▅█▇▁ beats 65/246	easy · avg 70%	▇▇▇ 7.4%	sharp · at its level
Decision Index 0.4.1: ARC-Challenge (accuracy) submitted	78.2% ✎	▂▂▂▂▂▃▁▃▄▆█▆ beats 68/246	very easy · avg 90%	▇▇▇ 7.2%	sharp
Decision Index 0.4.1: BBH (accuracy) submitted	52.5% ✎	▄▆▄▄█▇▇▄▃▁▁▁ beats 69/246	easy · avg 66%	▇▇ 6.2%	at its level
Decision Index 0.4.1: ARC-Easy (accuracy) submitted	90.7% ✎	▁▁▂▁▂▂▃▂▃▆█▅ beats 69/246	very easy · avg 96%	▇▇ 4.9%	sharp · maxed out · noisy
Decision Index 0.4.1: WinoGrande (accuracy) submitted	69.4% ✎	█▅█▅▄█▆▄▃▃▁▁ beats 78/246	easy · avg 78%	▇ 3.4%	at its level
Decision Index 0.4.1: GSM8K (accuracy (4- and 10-choice)) submitted	37.6% ✎	▁··▄█▆▆▆▅▅▃▁ beats 101/246	medium · avg 45%	▇ 2.6%	sharp · at its level
Decision Index 0.4.1: GPQA Diamond (accuracy) submitted	34.2% ✎	▂▆▅▅▆█▆▃▁··▁ beats 80/246	medium · avg 39%	▇ 2.4%	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	58.5% ✎	▁▁▂█▅▄▁▁▁▁▁▁ beats 114/246	medium · avg 60%	▇ 1.0%	blunt · at its level
Decision Index 0.4.1: CRUXEval (accuracy) submitted	30.2% ✎	██▄▃▃▄▁▁···▁ beats 2/246	medium · avg 58%	▇ 1.0%	at its level · noisy
Decision Index 0.4.1: CLadder (accuracy) submitted	58.9% ✎	▁███▆▅▃▃▁▁▁▁ beats 59/246	easy · avg 71%	▇ 0.9%	at its level
Decision Index 0.4.1: SATA-Bench (case exact accuracy) submitted	22.5% ✎	▁▁▁▁▁▁▁▂▅█▂▁ beats 87/246	hard · avg 22%	▇ 0.9%	typical
Decision Index 0.4.1: ChessBench (accuracy) submitted	8.9% ✎	▄▇▄▄▄▆▄█▄▁▁▁ beats 45/246	hard · avg 16%	▇ 0.4%	blunt
Decision Index 0.4.1: HLE (accuracy) submitted	12.8% ✎	▁▁▂▃▆█▇▅▅▃▁▁ beats 135/246	very hard · avg 13%	▇ 0.0%	blunt

other tests2 not in a task · 6.7% of its score

test	result	vs the others in its run	difficulty	counts	why
Decision Index 0.4.1: private same skills (equated points)	38.7200 points ✎	▂▂▂▃▃▃▄▅▅▇█▃ beats 92/247	avg 41.206 points	▇▇ 4.3%	blunt
Decision Index 0.4.1: private new domains (equated points)	15.3500 points ✎	▃█▃▃▂▆▃▃▄▄▄▄ beats 59/247	avg 38.861 points	▇ 2.4%	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/Kyunghwan-kim ↗

$ 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		= 44.3

not counted (2)

Decision Index 0.4.1	Decision Index 0.4.1: RouterBench (selected quality (quality objective)) [correct]	71.2%	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]	38.9%	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