Skip to content

Standard One 3B SH (v2)

Standard ThinkingDecision modelOpen weights

Standard One 3B SH (v2) by Standard Thinking is a decision model: it answers typed questions with a probability for each allowed answer instead of generating text. It scores 39.9 on the Jev Decision Index (rank 83).

Pricing

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

Specifications

Provider
Standard Thinking
Parameters
4B
Released
Oct 2026
Open weights
Yes
Question types
Choice (one of N options), Score (ordered rubric), Yes / no
Mechanism
Decoder LM with a decision readout
Backbone
StandardThinking/StandardOne-3B
Jev-compatible request
Yes
Runs on CPU
No
License
apache-2.0
Hugging Face
StandardThinking/StandardOne-3B-SH

Benchmarks

Jev Decision IndexECE 0.101
39.9

Jev Decision Index: public 42.9, ECE 0.101, Brier 0.475. 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 0.3 kit (release-v3), self-run
51.5

Speed & latency

Median latency per call (Decision Index)
76 ms

Compare Standard One 3B SH (v2) with…

FAQ

How much does Standard One 3B SH (v2) cost?

Standard One 3B SH (v2) has no hosted price listed; its weights are open, so the cost is the compute you run it on.

What questions can Standard One 3B SH (v2) answer?

Standard One 3B SH (v2) answers choice (one of n options), score (ordered rubric), yes / no questions, returning a probability for each allowed answer.

Is Standard One 3B SH (v2) well calibrated?

On the Jev Decision Index its expected calibration error (ECE) is 0.101; below about 0.05, a stated 80% confidence is right roughly 80% of the time.

Is Standard One 3B SH (v2) open weight, and can it run on a CPU?

Yes, Standard One 3B SH (v2) 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.