Lev1 3.7B
Lev1 3.7B by theaviv is a decision model: it answers typed questions with a probability for each allowed answer instead of generating text. It scores 38.6 on the Jev Decision Index (rank 88).
Pricing
No hosted price is listed. The weights are open: the cost is your own compute.
Specifications
- Provider
- theaviv
- 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
- IFM/K2-Horizon-3.7B
- Jev-compatible request
- No
- Runs on CPU
- No
- License
- apache-2.0
- Hugging Face
- theaviv/Lev1-3.7B
Benchmarks
- Jev Decision IndexECE 0.058
- 38.6
Jev Decision Index: public 42.7, ECE 0.058, Brier 0.447. 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
- 42.3
Speed & latency
Compare Lev1 3.7B with…
FAQ
How much does Lev1 3.7B cost?
Lev1 3.7B has no hosted price listed; its weights are open, so the cost is the compute you run it on.
What questions can Lev1 3.7B answer?
Lev1 3.7B answers choice (one of n options), score (ordered rubric), yes / no questions, returning a probability for each allowed answer.
Is Lev1 3.7B well calibrated?
On the Jev Decision Index its expected calibration error (ECE) is 0.058; below about 0.05, a stated 80% confidence is right roughly 80% of the time.
Is Lev1 3.7B open weight, and can it run on a CPU?
Yes, Lev1 3.7B 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, Hugging Face). Pricing is per 1M tokens (USD); confirm with the provider before production use.