system-one-gemma
system-one-gemma by akash-kamat is a decision model: it answers typed questions with a probability for each allowed answer instead of generating text. It scores 1.3 on the Jev Decision Index (rank 156).
Pricing
No hosted price is listed. The weights are open: the cost is your own compute.
Specifications
- Provider
- akash-kamat
- Parameters
- 268M
- 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
- google/gemma-3-270m
- Jev-compatible request
- No
- Runs on CPU
- Yes
- License
- —
Benchmarks
- Jev Decision IndexECE 0.239
- 1.3
Jev Decision Index: public 5.2, ECE 0.239, Brier 0.799. Lower ECE means its probabilities can be taken at face value.
Speed & latency
Compare system-one-gemma with…
FAQ
How much does system-one-gemma cost?
system-one-gemma has no hosted price listed; its weights are open, so the cost is the compute you run it on.
What questions can system-one-gemma answer?
system-one-gemma answers choice (one of n options), score (ordered rubric), yes / no questions, returning a probability for each allowed answer.
Is system-one-gemma well calibrated?
On the Jev Decision Index its expected calibration error (ECE) is 0.239; below about 0.05, a stated 80% confidence is right roughly 80% of the time.
Is system-one-gemma open weight, and can it run on a CPU?
Yes, system-one-gemma 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). Pricing is per 1M tokens (USD); confirm with the provider before production use.