jeff
jeff by logan-markewich is a decision model: it answers typed questions with a probability for each allowed answer instead of generating text. It scores 7.7 on the Jev Decision Index (rank 146).
Pricing
No hosted price is listed. The weights are open: the cost is your own compute.
Specifications
- Provider
- logan-markewich
- Parameters
- 576M
- Released
- Sep 2026
- Open weights
- Yes
- Question types
- Choice (one of N options), Score (ordered rubric), Yes / no
- Mechanism
- Span encoder (GLiNER family)
- Backbone
- knowledgator/gliformer-large-v1
- Jev-compatible request
- No
- Runs on CPU
- Yes
- License
- —
Benchmarks
- Jev Decision IndexECE 0.097
- 7.7
Jev Decision Index: public 8.3, ECE 0.097, Brier 0.682. Lower ECE means its probabilities can be taken at face value.
Speed & latency
Compare jeff with…
FAQ
How much does jeff cost?
jeff has no hosted price listed; its weights are open, so the cost is the compute you run it on.
What questions can jeff answer?
jeff answers choice (one of n options), score (ordered rubric), yes / no questions, returning a probability for each allowed answer.
Is jeff well calibrated?
On the Jev Decision Index its expected calibration error (ECE) is 0.097; below about 0.05, a stated 80% confidence is right roughly 80% of the time.
Is jeff open weight, and can it run on a CPU?
Yes, jeff 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.