whatmodel.sh / decision / Bespoke Nimble 9B v2

Bespoke Nimble 9B v2

+ contribute

Bespoke Nimble 9B v2 is an open-weights decision model from Bespoke Labs. Its published size is 9.65B parameters. Licence: apache-2.0.

Catalogue identity notes

Exact Decision Index tested engine nimble-v2; reported model Bespoke Nimble 9B v2. Type: LoRA; technique: autoregressive. Implementation: https://github.com/bespokelabsai/nimble. No 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. Reviewed Ollaya local route added 2026-10-31: nimble:9b, manifest fb7e7f742e34636e853f91c467b9463e3155dbb08cabc5c49906d52a5750487d, matching the independent accuracy-run manifest. Native family implementation; published runtime precision label F32. Memory is measured peak GPU increase (20782 MiB, rounded upward to 20.3 GiB) on RTX 5090 during the short five-question triage workload, not download size or a maximum-context guarantee. Runtime installation is required; no fresh execution performed here. Historical benchmark adapter and input protocol still vary by result.

Model information sources

· Weights licence ↗ · Parameter count ↗

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

50.3±1.1#203 of 355 decision modelsfrom 56 benchmark results across 2 sources

$ how it compares

The decision models ranked either side of Bespoke Nimble 9B v2, out of 357 with a score, then more from Bespoke Labs.

#	model	score	where	RAM	released
203	OneJev-4B (JevBench mini) OmniJev	51±2	source code
204	JPT-4B Qwen3.5-4B LoRA kirp	50±1	model files
205	Bespoke Nimble 9B v2 Bespoke Labs	50±1	local	~20GB
206	JevK5 v0.3 alibiserikbay	50±1	model files
207	Span-01 Pro Respan	50±9	api price unknown		2026-09-24
	more from Bespoke Labs
15	Bespoke Nimble v4 Max (27B) Bespoke Labs	70±1	model files		2026-10
135	Bespoke Nimble 9B v3 Bespoke Labs	58±1	local	~21GB	2026-10-06
172	Bespoke-Nimble-9B (DecisionBench 594dfdcf) Bespoke Labs	54±2	model files

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

$ what it's good at

tools & automation▁▃▁▁▁▂▃▄▆██▆ #164 of 250 · 6.7% of its score · 6 tests

Beats LiquidAI d1-3B, vLLM-SR Decision 2.0 Nox 4B, Decider 4B here, though they rank higher overall. Trails JevK5 Lite, Gevva E2B, Amber 0.4B 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	91.6% ✎	▁▂▁▁▁▂▂▄▅▆█▁ beats 84/246	very easy · avg 92%	▇ 2.1%	sharp · maxed out · noisy
Decision Index 0.4.1: API-Bank (accuracy) submitted	80.5% ✎	▁▁▁▁▁▁▁▂▃▂█▂ beats 120/236	easy · avg 70%	▇ 2.0%	sharp · at its level · noisy
Decision Index 0.4.1: When2Call (accuracy) submitted	72.7% ✎	▁▁▁▂▃▅▅▆█▇█▃ beats 105/246	easy · avg 72%	▇ 0.9%	at its level · noisy
Decision Index 0.4.1: Home appliance simulator (case exact accuracy) submitted	2.3% ✎	█▃▁▃▄▄▆▄▁▂▁▃ beats 54/243	medium · avg 37%	▇ 0.8%	sharp · at its level · noisy
Decision Index 0.4.1: ToolRet (nDCG@10) submitted	58.0% ✎	▁▁▁▁▁·▁▁▁▄█▁ beats 54/246	medium · avg 60%	▇ 0.8%	blunt · at its level
Decision Index 0.4.1: RouterBench (selected quality (quality objective)) submitted	78.3% ✎	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▁▁▁·▁▁▁▁▃▃█▅ #174 of 258 · 5.1% of its score · 7 tests

Beats Solar Decide, JevK5 9B v0.3.3, Aplomb 1 here, though they rank higher overall. Trails Mola-68M, Bosun v3.1 1.7B LoRA, Dinah-1 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	28.1% ✎	▁▁▁▁▁▁▁▃▂▄▇█ beats 39/246	medium · avg 38%	▇ 1.8%	blunt · at its level
Decision Index 0.4.1: Amazon ESCI (macro-F1) submitted	45.5% ✎	▁▁▁▁▁▁▂▂▃▃█▅ beats 85/246	medium · avg 47%	▇ 1.7%	at its level
Decision Index 0.4.1: HoVer (accuracy) submitted	62.9% ✎	▆▆▇▅▄▄▃▇▄█▆▄ beats 69/246	easy · avg 74%	▇ 0.5%	blunt · at its level
Decision Index 0.4.1: PhishNChips phishing decisions (accuracy) submitted	65.6% ✎	█▅▅▄▃▂▁▁▁▁·▁ beats 108/245	easy · avg 70%	▇ 0.4%	at its level · noisy
Decision Index 0.4.1: CLINC150+OOS (macro-F1) submitted	30.8% ✎	▁▁▁▁▁▃▁▂▃▆█▄ beats 18/238	easy · avg 83%	▇ 0.4%	sharp · noisy
Decision Index 0.4.1: BANKING77 (macro-F1) submitted	69.4% ✎	▁·▁▁▁▁▁▁▂█▇█ beats 38/238	easy · avg 79%	▇ 0.3%	noisy
Decision Index 0.4.1: SGD/SGD-X (macro-F1) submitted	59.9% ✎	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▂▂▂▃▂▅▆▃█▆▄▃ #141 of 256 · 20% of its score · 9 tests

Beats Solar Decide, JoshuaSP diffgemma (decision readout), PurpleMIST-Flash-1.0 here, though they rank higher overall. Trails Decider 2B, Ryotide Qwen v0.3.1 (decision readout), Sieve 2B v3 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	85.0% ✎	▂▄▂▂▄▇▃▇█▅▃▁ beats 113/246	very easy · avg 85%	▇▇ 5.4%	sharp
Decision Index 0.4.1: NLI4CT (macro-F1) submitted	76.3% ✎	▁▁▁▂▂▃▃▄▇▇█▄ beats 111/246	easy · avg 74%	▇▇ 4.5%	at its level
Decision Index 0.4.1: ANLI (macro-F1) submitted	68.2% ✎	▁▂▃▅▅█▂····▁ beats 145/246	medium · avg 60%	▇▇ 4.3%	at its level
Decision Index 0.4.1: iSarcasmEval (Sarcasm F1 · track A, English) submitted	53.7% ✎	▁·▁·▁▁▂▄▅█▇▄ beats 149/246	medium · avg 44%	▇ 1.7%	at its level
Decision Index 0.4.1: ContractNLI (macro-F1) submitted	70.3% ✎	▁▁▁▁▁▁▁▂▂▇▇█ beats 76/246	easy · avg 71%	▇ 0.9%	at its level · noisy
Decision Index 0.4.1: VAST (macro-F1) submitted	39.1% ✎	▁▁▁▂▅▄▄▂▄▆█▄ beats 26/246	medium · avg 62%	▇ 0.8%	at its level
Decision Index 0.4.1: ACOS (per-review F1) submitted	23.1% ✎	▁··▂▃▃▅▇█▇▃▃ beats 131/246	hard · avg 22%	▇ 0.8%	noisy
Decision Index 0.4.1: RAGTruth (F1 on hallucinated class) submitted	65.1% ✎	▁▁▁▁▁▁▁▁▁▃▆█ beats 77/246	easy · avg 66%	▇ 0.7%	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	65.5% ✎	▁···▁▁▁▁▃▆█▃ beats 31/246	easy · avg 84%	▇ 0.6%	noisy

arts & human taste▁▇▄▃▅▅█▅▄▂▁▁ #133 of 247 · 7.4% of its score · 6 tests

Beats OneJev1.6-4B, Nace.AI Drex v1.5, Cerno 4B v1.1 here, though they rank higher overall. Trails vLLM-SR vela3-fast-vision (Vela 3.0, 0.25B), Crivo-400M, Amber 0.4B 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	55.7% ✎	▁▁▂▃▃▃▇▅█▅▃▁ beats 77/246	medium · avg 61%	▇ 3.2%	at its level
Decision Index 0.4.1: Humicroedit (accuracy) submitted	57.5% ✎	▁·▁▃▂▃▅█▄▂▁▁ beats 72/246	medium · avg 60%	▇ 1.5%	blunt · at its level
Decision Index 0.4.1: BPoMP (accuracy) submitted	86.7% ✎	▁█▄▄▅█▆▅▄▁▁▁ beats 114/246	easy · avg 83%	▇ 1.1%	sharp · noisy
Decision Index 0.4.1: cfcolor (accuracy) submitted	54.3% ✎	▁········▁▇█ beats 51/246	medium · avg 60%	▇ 0.8%	blunt · at its level
Decision Index 0.4.1: POP909-CL (accuracy) submitted	14.6% ✎	▄▅▅▃▆▅▃▅▂▅▃█ beats 119/237	hard · avg 21%	▇ 0.5%	sharp · noisy
Decision Index 0.4.1: Habermas Machine (accuracy) submitted	40.4% ✎	▁▂▄█▇▅▁▁▄▄▁▁ beats 71/246	medium · avg 47%	▇ 0.2%	blunt · at its level

knowledge & reasoning▁▂▂▂▂▃▃▅▅▆█▂ #129 of 263 · 43% of its score · 19 tests

Beats Winnow-12B (DecisionBench b6ac22b0), Cerno 4B v1.1, Amber Hybrid 4B (decision readout) here, though they rank higher overall. Trails djev 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: MMLU-Pro (accuracy) submitted	51.7% ✎	▄▄▃▃▂▄▅▇██▅▁ beats 120/246	medium · avg 46%	▇▇▇ 7.9%	sharp · at its level
Decision Index 0.4.1: MMLU (accuracy) submitted	77.2% ✎	▁▄▄▃▃▂▅▇▅█▇▁ beats 133/246	easy · avg 70%	▇▇ 6.1%	sharp · at its level
Decision Index 0.4.1: BBH (accuracy) submitted	67.4% ✎	▄▆▄▄█▇▇▄▃▁▁▁ beats 119/246	easy · avg 66%	▇▇ 5.3%	at its level
Decision Index 0.4.1: ARC-Challenge (accuracy) submitted	94.1% ✎	▂▂▂▂▂▂▁▃▄▆█▆ beats 129/246	very easy · avg 90%	▇▇ 4.4%	sharp
Decision Index 0.4.1: WinoGrande (accuracy) submitted	75.6% ✎	█▅█▅▄█▆▄▃▃▁▁ beats 113/246	easy · avg 78%	▇ 2.7%	typical
Decision Index 0.4.1: GSM8K (accuracy (4- and 10-choice)) submitted	43.3% ✎	▁··▄█▆▆▆▅▅▃▁ beats 123/246	medium · avg 45%	▇ 2.6%	sharp · at its level
Decision Index 0.4.1: ARC-Easy (accuracy) submitted	98.1% ✎	▁▁▂▁▂▂▂▂▃▆█▅ beats 133/246	very easy · avg 96%	▇ 2.5%	sharp · maxed out · noisy
Decision Index 0.4.1: GPQA Diamond (accuracy) submitted	43.9% ✎	▂▆▅▅▆█▆▃▁··▁ beats 142/246	medium · avg 39%	▇ 2.3%	blunt · at its level

11 more tests

test	result	vs the others in its run	difficulty	counts	why
Ollaya · PubMedQA	75.6% ✎	▃▂▃▃·▂▃▃▆█▆▂ beats 21/25	easy · avg 69%	▇ 1.8%	at its level
Ollaya · BoolQ	85.3% ✎	▁··▃▃▁▄▁▃█▇▄ beats 14/25	easy · avg 84%	▇ 1.4%	typical
Decision Index 0.4.1: MuSR (accuracy) submitted	56.4% ✎	▁▁▂█▅▄▁▁▁▁▁▁ beats 89/246	medium · avg 60%	▇ 1.0%	blunt · at its level
Decision Index 0.4.1: SATA-Bench (case exact accuracy) submitted	23.5% ✎	▁▁▁▁▁▁▁▂▅█▂▁ beats 94/246	hard · avg 22%	▇ 0.9%	typical
Decision Index 0.4.1: CRUXEval (accuracy) submitted	56.3% ✎	██▄▃▃▄▁▁···▁ beats 134/246	medium · avg 58%	▇ 0.9%	at its level · noisy
Decision Index 0.4.1: CLadder (accuracy) submitted	66.9% ✎	▁███▆▅▃▃▁▁▁▁ beats 111/246	easy · avg 71%	▇ 0.8%	at its level
Ollaya · SQuAD 2 answerability	82.3% ✎	▁▁▃▁▇█▃▃▄▃·▁ beats 18/25	easy · avg 80%	▇ 0.7%	blunt
Ollaya · VitaminC verification	76.8% ✎	▁···▃·▄█▃··▁ beats 12/25	easy · avg 78%	▇ 0.7%	blunt
Decision Index 0.4.1: ChessBench (accuracy) submitted	13.6% ✎	▄▇▄▄▄▆▄█▄▁▁▁ beats 97/246	hard · avg 16%	▇ 0.5%	blunt
Ollaya · MultiNLI	82.6% ✎	▂··▄██▂▆▆█▆▂ beats 8/25	very easy · avg 87%	▇ 0.4%	blunt
Decision Index 0.4.1: HLE (accuracy) submitted	8.6% ✎	▁▁▃▃▆█▇▅▅▃▁▁ beats 7/246	very hard · avg 13%	▇ 0.0%	blunt

workflows▁▁▁▂▁▂▂▆█▄▁▁ #69 of 114 · 11% of its score · 9 tests

Beats Clef-omni, baselabs joey-1 (2B), ZTC-Judge-9B (DecisionBench 05d154a5) here, though they rank higher overall. Trails Solar Decide Flash, InternLM Intern-Decision Qwen3.5-4B, StartLux-Decision v1.1 0.8B here, though they rank lower overall.

test	result	vs the others in its run	difficulty	counts	why
Ollaya · Typed decisions test	66.5% ✎	▄·▂▆▄▄▂·▄█▆▆ beats 13/22	medium · avg 64%	▇ 2.9%	at its level
Ollaya · PAWS paraphrase	80.0% ✎	▂▂▂▂·▃█▆·▂·▁ beats 11/25	easy · avg 81%	▇ 1.8%	typical
Ollaya · SummEval relevance	40.0% ✎	▁··▁▂▁▅··▃██ beats 11/25	medium · avg 41%	▇ 1.5%	at its level
Ollaya · MASSIVE German intent	80.9% ✎	▁····▁·▂▃▆█▃ beats 10/23	easy · avg 84%	▇ 1.5%	sharp
Ollaya · Aegis 2 safety	81.2% ✎	▂·▁▁▂·▃▃▂█▃▁ beats 19/25	easy · avg 77%	▇ 1.5%	typical
Ollaya · SummEval consistency	84.7% ✎	▁·▁▁··▁▁▁▁▄█ beats 23/25	easy · avg 75%	▇ 0.9%	sharp · noisy
Ollaya · MASSIVE English intent	84.0% ✎	▁·······▂▃█▃ beats 9/23	very easy · avg 86%	▇ 0.8%	noisy
Ollaya · HelpSteer 2 scoring	34.1% ✎	▁▁··▁·▁▃▇█▅▄ beats 7/23	medium · avg 38%	▇ 0.6%	blunt · at its level

1 more test

test	result	vs the others in its run	difficulty	counts	why
Ollaya · Civil Comments	84.3% ✎	▂▂▂·▂▅▅▆▃▅·█ beats 9/23	very easy · avg 86%	▇ 0.1%	blunt · noisy

other tests2 not in a task · 6.3% of its score

test	result	vs the others in its run	difficulty	counts	why
Decision Index 0.4.1: private same skills (equated points)	38.9100 points ✎	▂▂▂▃▃▃▄▅▅▇█▃ beats 93/247	avg 41.206 points	▇▇ 4.0%	blunt
Decision Index 0.4.1: private new domains (equated points)	46.7800 points ✎	▃█▃▃▂▆▃▃▄▄▄▄ beats 147/247	avg 38.861 points	▇ 2.2%	blunt

$ where to run it

Other ways to run it

Run it as published · needs 20.3 GB, NVIDIA only · from github.com/ollaya-dev
ollaya run nimble:9b --preset triage "The customer requests a refund."

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 matching Ollaya runtime manifest and published short-workload GPU memory measurement were reviewed separately; see local variants and identity_note. The command was not executed in this review. Source · checked 2026-10-31.

Model repository and documentation: huggingface.co/bespokelabs ↗

$ how the score was worked out · Decision Index 0.4.1 83%, Ollaya · original runs 17%

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	▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇ 83%	CC BY 4.0
Ollaya · original runs	14	▇▇▇▇▇ 17%	Apache-2.0
prior · pull toward the middle		▇ 1%
weighted average		= 50.3

not counted (2)

Decision Index 0.4.1	Decision Index 0.4.1: RouterBench (selected quality (quality objective)) [correct]	78.3%	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]	59.9%	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