The LLM benchmark aggregator.

Bespoke Nimble 9B

Other · 9B

Strongest in Routing & classification (#104 of 234), weakest in Calibration (#193 of 234). Above par in 1 of 3 scopes. Among the models it meets almost everywhere, it finishes behind Baselabs mammoth-1 and KnowLine 31B-Gen1 and ahead of Diffusion Jev and APUS OpenJev V1 9B.

§ 1 · Profile

What it is good at

Bars run from 50 — the average of the models each source lists — so right of the line is above par. The middle column is the gap to whoever leads that scope.

Decisions47.9−18#158/2341/1

Routing & classification52.2−19.5#104/2341/1

Calibration43.7−18#193/2341/1

§ 2 · Head to head

What it beats, and what beats it

The same models turn up scope after scope. Counted once: where both were placed, who finished higher. Bars run right for Bespoke Nimble 9B, left for the other.

CLM 8Bwon 3 · lost 0

Laya multilingualwon 3 · lost 0

Julia 1won 3 · lost 0

Tacet Sonatawon 3 · lost 0

Raw Qwen3 1.7B direct logitswon 3 · lost 0

Raw Qwen3 0.6B direct logitswon 3 · lost 0

ClassOne Qwen 3.5 2Bwon 3 · lost 0

Deem 0.8B v1won 3 · lost 0

Mirrorwon 3 · lost 0

ClassOne Gemma 4 E2Bwon 3 · lost 0

openJev Verdictwon 3 · lost 0

SimpleJevwon 3 · lost 0

Phocinae Largha 150M V1won 3 · lost 0

Quyet-1.0-Tiny (Chinh Nguyen, mmBERT-small encoder (16 layers kept) with fixed heads)won 3 · lost 0

Quyet 1.0 Smallwon 3 · lost 0

Layawon 3 · lost 0

kev 0.5Bwon 3 · lost 0

Liquid AI d1-omni-600Mwon 3 · lost 0

Vega 0.8Bwon 3 · lost 0

Raw Qwen3 8B direct logitswon 3 · lost 0

Raw Phi-4 mini direct logitswon 3 · lost 0

System Onewon 3 · lost 0

ClassOne Qwen 3.5 4Bwon 3 · lost 0

Vega 4Bwon 3 · lost 0

ClassOne Gemma 4 E4Bwon 3 · lost 0

WaterSheepwon 3 · lost 0

Open Jev Deberta V3 Largewon 3 · lost 0

Raw Qwen3 4B Instruct direct logitswon 3 · lost 0

kev 0.6Bwon 3 · lost 0

OpenDecisionwon 3 · lost 0

Chon 1.0 12Bwon 3 · lost 0

Krino 2Bwon 3 · lost 0

Decisor 4Bwon 3 · lost 0

kev 4Bwon 3 · lost 0

Decider 2Bwon 3 · lost 0

ClassOne Qwen 3.5 9Bwon 3 · lost 0

Verdict Smallwon 2 · lost 1

Quyet 1.0 Small ENwon 2 · lost 1

openJev Verdict 1.4won 2 · lost 1

Decision Fastwon 2 · lost 1

Jeetu Decider 2Bwon 2 · lost 1

Gutsy 0.8B v0.3won 2 · lost 1

Watt Flash 0.1won 2 · lost 1

Vega Uwon 2 · lost 1

Jul fastwon 2 · lost 1

Alibaba GTE Reranker ModernBERT-basewon 2 · lost 1

Smalljev semantic-v9won 2 · lost 1

PurpleMIST Mini 1.0won 2 · lost 1

Qwen3 Reranker 4Bwon 2 · lost 1

Jez Small 1.0.0won 2 · lost 1

Lev 350Mwon 2 · lost 1

JADEwon 2 · lost 1

Certo v1won 2 · lost 1

Laya typed-decisionswon 2 · lost 1

Mixedbread mxbai-rerank-base-v2won 2 · lost 1

BAAI bge-reranker-v2-m3won 2 · lost 1

D1A on Gemma 4 E2Bwon 2 · lost 1

Nemotron Diffusion 8Bwon 2 · lost 1

Open-Jev 2Bwon 2 · lost 1

Jiwo 0.8bwon 2 · lost 1

Open Jev JSON Canvaswon 2 · lost 1

DecisionNode Edge Mini 0.9.2won 2 · lost 1

OpenSourceJevwon 2 · lost 1

Malkuth 2Bwon 2 · lost 1

EXAONE-4.0-1.2B-JEV v0.3won 2 · lost 1

kev 8Bwon 2 · lost 1

Open Alternative Jevwon 2 · lost 1

Liquid AI d1-3Bwon 2 · lost 1

ZeroEntropy zerank-2won 2 · lost 1

Spark S1 4B V6won 2 · lost 1

D1A on Gemma 4 E4Bwon 2 · lost 1

DecisionNode Edge 0.9.2won 2 · lost 1

Arbiter 4Bwon 2 · lost 1

RYOTIDE Qwen9won 2 · lost 1

APUS OpenJev V1 9Bwon 2 · lost 1

Diffusion Jevwon 2 · lost 1

Jez Large 1.0.0won 1 · lost 2

Nomawon 1 · lost 2

Imajev 2Bwon 1 · lost 2

Standard One 3B SHwon 1 · lost 2

ArseneLupin v1.1won 1 · lost 2

Standard One 3B SH v4won 1 · lost 2

TypeCastLM 1.4.0won 1 · lost 2

SemIfwon 1 · lost 2

local-jev Qwen3.5-4Bwon 1 · lost 2

Decision 2Bwon 1 · lost 2

Jev Localwon 1 · lost 2

Compass 0.2.0won 1 · lost 2

Messier One v0.2won 1 · lost 2

Uniphore 4B Deciderwon 1 · lost 2

Typecastlmwon 1 · lost 2

CoCo Decision 4Bwon 1 · lost 2

Pjev 2Bwon 1 · lost 2

Jebadiah 9B v2won 1 · lost 2

RYOTIDE Qwenwon 1 · lost 2

Iota Decide 4Bwon 1 · lost 2

JEV Qwen3.5-9B Basewon 1 · lost 2

Decider 4Bwon 1 · lost 2

Decision 4B v1.1won 1 · lost 2

Quyet 1.0won 1 · lost 2

Clef Omniwon 1 · lost 2

Manchego v2.1won 1 · lost 2

Standard One 8Bwon 1 · lost 2

Decision 4Bwon 1 · lost 2

Wald 4Bwon 1 · lost 2

reflex 4Bwon 1 · lost 2

Verdin 4Bwon 1 · lost 2

Qwen3.5-9B Jev-like data-mix v2won 1 · lost 2

Standard One 8B SHwon 1 · lost 2

Jiwo 4bwon 1 · lost 2

Decision 4B v1.2won 1 · lost 2

Levwon 1 · lost 2

Kahn1 4Bwon 1 · lost 2

Metask Jev 4Bwon 1 · lost 2

Malkuth 4Bwon 1 · lost 2

Hopperwon 1 · lost 2

Blink 4Bwon 1 · lost 2

Decision 2.0 Lux 9Bwon 1 · lost 2

Decisio v0.9.0 on Qwen3.6-35B-A3Bwon 1 · lost 2

JEB 35B A3Bwon 1 · lost 2

Imajev 4Bwon 1 · lost 2

Jqvwon 1 · lost 2

Imajev 9Bwon 1 · lost 2

janus 4Bwon 1 · lost 2

GLiClass v4 9Bwon 1 · lost 2

JevK5 v0.2.0won 1 · lost 2

Seb 9Bwon 1 · lost 2

Clef Flashwon 1 · lost 2

JPT 4Bwon 1 · lost 2

deck-4B v1.0won 1 · lost 2

Jebadiah 9B v2.1won 1 · lost 2

Jeffan 4Bwon 1 · lost 2

JevK5 v0.3won 1 · lost 2

Decider-4b v2won 1 · lost 2

Plumb 4Bwon 1 · lost 2

Messier One v0.4won 0 · lost 3

Pplx Decider Flash V1.2won 0 · lost 3

Open-Jev 9Bwon 0 · lost 3

Seb 9B V2won 0 · lost 3

Shisa De 1won 0 · lost 3

Arcbright Elf 1 12Bwon 0 · lost 3

Metask rain 4Bwon 0 · lost 3

Bespoke Nimble 9B v3won 0 · lost 3

Decider 35B A3Bwon 0 · lost 3

KnowLine 4B-Gen5.5won 0 · lost 3

Bev / Bonsai 27Bwon 0 · lost 3

Amazon Strands Decider 26B-A4Bwon 0 · lost 3

Tern 35B A3Bwon 0 · lost 3

LitJevwon 0 · lost 3

Open-Jev 27B v1.1won 0 · lost 3

Aplomb 1won 0 · lost 3

Chon 1.0 E4Bwon 0 · lost 3

NInfer Qwen3.8-27Bwon 0 · lost 3

Respan Span-02won 0 · lost 3

Hopper 12B trainedwon 0 · lost 3

Torchcast Decision 4B v1.1won 0 · lost 3

Iota Decide 4B V0.3won 0 · lost 3

Xevn 4Bwon 0 · lost 3

SPX CD Omniwon 0 · lost 3

Philly Decider v2won 0 · lost 3

Chon 1.0 12B Fastwon 0 · lost 3

vLLM-SR d3-miniwon 0 · lost 3

Wald 4B v2.1won 0 · lost 3

Jebadiah 27Bwon 0 · lost 3

decider chat on Gemma-4-31B-itwon 0 · lost 3

JevOnewon 0 · lost 3

Baselabs elephant-1won 0 · lost 3

decisio v0.9.0 on gemma-4-12B-itwon 0 · lost 3

Decider-12b v2won 0 · lost 3

decider-12b v1, stock Gemma-4-12B-itwon 0 · lost 3

NInfer Qwen3.8-Flash-Next mixedwon 0 · lost 3

Decision 2.0 Vega 27Bwon 0 · lost 3

Winnow-12B Q8won 0 · lost 3

Blink v0.3 26B-A4Bwon 0 · lost 3

GEV 26B Decidewon 0 · lost 3

Xor 26B-A4Bwon 0 · lost 3

Cygnetwon 0 · lost 3

Torchcast Decision 12Bwon 0 · lost 3

decisio v0.8.0 on gemma-4-12B-itwon 0 · lost 3

JPT 35B A3Bwon 0 · lost 3

Jev Omniwon 0 · lost 3

AutoJev 27Bwon 0 · lost 3

Bobcat Flash 1.2won 0 · lost 3

Torchcast Decision 4B v1.2won 0 · lost 3

FrontiersMind Lumma-fev-27Bwon 0 · lost 3

swanOnewon 0 · lost 3

RSI-Jev v6.1-VL 27Bwon 0 · lost 3

Nagi-ENORMOUS 27Bwon 0 · lost 3

Surogate Rune 26B-A4B v3won 0 · lost 3

JEV 27Bwon 0 · lost 3

OpenJevwon 0 · lost 3

H2O-Lightning-4B v1.1won 0 · lost 3

SimpleJev Qwen3.8-27Bwon 0 · lost 3

Decisio 12B tuned commitwon 0 · lost 3

Clefwon 0 · lost 3

René-1 31Bwon 0 · lost 3

Aleria Judge 1.1.1won 0 · lost 3

Deck 31Bwon 0 · lost 3

Metask Jev Rain 12B Trainedwon 0 · lost 3

Wald 26B-A4Bwon 0 · lost 3

Torchcast Decision 4B v1.3won 0 · lost 3

KnowLine 26B-A4B-Gen2.5won 0 · lost 3

Eikos 27Bwon 0 · lost 3

KnowLine 35B-A3B-Gen1.5won 0 · lost 3

Bobcat 1.3won 0 · lost 3

Metask Jev Rain 31B Trainedwon 0 · lost 3

metask-jev-rain-31B-trained service v3won 0 · lost 3

StartLux-Decision v1.1 27B AutoThinkwon 0 · lost 3

Lychee 27Bwon 0 · lost 3

decisio v0.8.0 on gemma-4-31B-itwon 0 · lost 3

Kev 27Bwon 0 · lost 3

Vera Largewon 0 · lost 3

Torchcast Decision 27Bwon 0 · lost 3

Pplx Decider V1.2won 0 · lost 3

Turion Jev 31Bwon 0 · lost 3

Nimble V4won 0 · lost 3

AJev lm6qwon 0 · lost 3

Sieve 31B V2won 0 · lost 3

Quyet 1.0 Largewon 0 · lost 3

Jeff 1.0 Largewon 0 · lost 3

Jeff 1.1won 0 · lost 3

Jeffan 31Bwon 0 · lost 3

Arcbright Elf 1 31Bwon 0 · lost 3

DOPEST 1.0 31Bwon 0 · lost 3

Basal Omni 1.0 Ultrawon 0 · lost 3

vLLM-SR d3won 0 · lost 3

Perplexity Decider v1.1 27Bwon 0 · lost 3

vLLM-SR d3-turbowon 0 · lost 3

StartLux-Decision v1.1 27B Instantwon 0 · lost 3

H2O-Lightning-31B v1.0.7won 0 · lost 3

Fastino GLiDE 1.2won 0 · lost 3

KnowLine 31B-Gen1won 0 · lost 3

Baselabs mammoth-1won 0 · lost 3

§ 3 · Sources

Where the numbers come from

1 publication, 2 figures. Every one links to the page it was read from.

JevBench ↗2 measuresread 2026-11-10

JevBench · Intelligence 43.1%JevBench · Calibration 69.9%

Badge

[![PublicAI Index](https://publicai.io/model-index/badge?model=bespoke-nimble-9b)](https://publicai.io/model-index/m/bespoke-nimble-9b)