Skip to content

Bespoke Nimble 9B v3

Bespoke LabsDecision modelOpen weights

Bespoke Nimble 9B v3 by Bespoke Labs is a decision model: it answers typed questions with a probability for each allowed answer instead of generating text. It scores 54.7 on the Jev Decision Index (rank 36).

Pricing

No hosted price is listed. The weights are open: the cost is your own compute.

Specifications

Provider
Bespoke Labs
Parameters
10B
Released
Sep 2026
Open weights
Yes
Question types
Choice (one of N options), Score (ordered rubric), Yes / no
Mechanism
Decoder LM with a decision readout
Backbone
Qwen/Qwen3.5-9B
Jev-compatible request
Yes
Runs on CPU
No
License
cc-by-nc-4.0
Hugging Face
bespokelabs/Bespoke-Nimble-9B-v3

Benchmarks

Jev Decision IndexECE 0.123
54.7

Jev Decision Index: public 57.2, ECE 0.123, Brier 0.370. Lower ECE means its probabilities can be taken at face value.

Self-reported by the publisher

The model's own claims, often on its own sets or a self-run copy of a public suite. Not independently verified and never ranked alongside measured results.

Jev Decision Index public suite (self-run)Self-reportedDecision Index public balanced skill, edition 0.2.1 kit (release-v2.1), self-run
56.9

Speed & latency

Median latency per call (Decision Index)
74 ms

Compare Bespoke Nimble 9B v3 with…

FAQ

How much does Bespoke Nimble 9B v3 cost?

Bespoke Nimble 9B v3 has no hosted price listed; its weights are open, so the cost is the compute you run it on.

What questions can Bespoke Nimble 9B v3 answer?

Bespoke Nimble 9B v3 answers choice (one of n options), score (ordered rubric), yes / no questions, returning a probability for each allowed answer.

Is Bespoke Nimble 9B v3 well calibrated?

On the Jev Decision Index its expected calibration error (ECE) is 0.123; below about 0.05, a stated 80% confidence is right roughly 80% of the time.

Is Bespoke Nimble 9B v3 open weight, and can it run on a CPU?

Yes, Bespoke Nimble 9B v3 has open weights. It needs a GPU to run at useful speed.

Data as of 2026-10-14. Source: Metatext decision-model catalogue, 2026-10-14 (Jev Decision Index, Benchmark Heaven JevBench, Hugging Face). Pricing is per 1M tokens (USD); confirm with the provider before production use.