All models

COMPARE

2 models, side by side.

What each one decides, how well calibrated it is, how fast it answers and what it costs to pull. Up to 4 at a time; the better value in each row is marked.

Propertybespoke-labs/bespoke-nimble-9bwfzyx/von
SummaryAn open Jev-style LoRA on Qwen3.5-9B from Bespoke Labs, trained on 2,676 contrastively curated examples to score the allowed answer tokens directly for enums, booleans and rubric levels. Recipe, data and a public benchmark suite are released with it.A ModernBERT-large encoder with an option-marker head. Premise and options are packed into one sequence and each option's marker is scored in a single bidirectional pass; version 1.2 makes the scoring order-invariant.
Decideschoice, noul, score, classify, routechoice, score, noul, classify
Architecturenimblevon
Fine-tuned fromqwen/qwen3.5-9banswerdotai/modernbert-large
Licenseapache-2.0apache-2.0
AvailabilityOpen weights + hosted APIOpen weights
Hosted byBespoke Labs—
Input price——
Decision accuracy90.1%63.9%
Calibration error0.0540.045
Valid action rate——
Median latency106 ms18 ms
p95 latency——
Evaluation suiteBespoke held-out set (324 examples)JevBench public standard tier
Latest version2026.091.2.0
VariantsSHA256SUMS—
Size of latest version184.3 MB2.9 GB
Files102
Downloads00
Stars00
Tagssystem-one, qwen, lora, curated-data, 9bsystem-one, encoder, modernbert, 395m
UpdatedSep 25, 2026Sep 25, 2026