Skip to content

Julia 1

Supersonic LabsDecision modelOpen weights

Julia 1 by Supersonic Labs is a decision model: it answers typed questions with a probability for each allowed answer instead of generating text. It scores 2.6 on the Jev Decision Index (rank 155).

Pricing

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

Specifications

Provider
Supersonic Labs
Parameters
141M
Released
Sep 2026
Open weights
Yes
Question types
Choice (one of N options), Score (ordered rubric), Yes / no
Mechanism
Encoder classifier
Backbone
jhu-clsp/mmBERT-small
Jev-compatible request
Yes
Runs on CPU
Yes
License
apache-2.0
Hugging Face
SupersonicLabs/Julia-1

Benchmarks

Jev Decision IndexECE 0.420
2.6

Jev Decision Index: public 4.2, ECE 0.420, Brier 0.987. Lower ECE means its probabilities can be taken at face value.

Speed & latency

Median latency per call (Decision Index)
6 ms

Compare Julia 1 with…

FAQ

How much does Julia 1 cost?

Julia 1 has no hosted price listed; its weights are open, so the cost is the compute you run it on.

What questions can Julia 1 answer?

Julia 1 answers choice (one of n options), score (ordered rubric), yes / no questions, returning a probability for each allowed answer.

Is Julia 1 well calibrated?

On the Jev Decision Index its expected calibration error (ECE) is 0.420; below about 0.05, a stated 80% confidence is right roughly 80% of the time.

Is Julia 1 open weight, and can it run on a CPU?

Yes, Julia 1 has open weights. It is small enough to run on a CPU.

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