jebadiah-4b-v1
jebadiah-4b-v1 by Frontier Infra is a decision model: it answers typed questions with a probability for each allowed answer instead of generating text.
Pricing
No hosted price is listed. The weights are open: the cost is your own compute.
Specifications
- Provider
- Frontier Infra
- Parameters
- 4B
- 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
- Qwen/Qwen3.5-4B-Base
- Jev-compatible request
- Yes
- Runs on CPU
- No
- License
- apache-2.0
- Hugging Face
- frontier-infra/jebadiah-4b-v1
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.
- Jevals suite 0.1.0, Banking77 (choice, 77-way)Self-reported
- 68.3
- Jevals suite 0.1.0, HelpSteer2 helpfulness (score)Self-reported
- 9.2
- Jevals suite 0.1.0, PubMedQA (noul)Self-reported
- 86.7
- Nimble public human-labelled subsets (13, macro accuracy)Self-reported
- 75.9
Compare jebadiah-4b-v1 with…
FAQ
How much does jebadiah-4b-v1 cost?
jebadiah-4b-v1 has no hosted price listed; its weights are open, so the cost is the compute you run it on.
What questions can jebadiah-4b-v1 answer?
jebadiah-4b-v1 answers choice (one of n options), score (ordered rubric), yes / no questions, returning a probability for each allowed answer.
Is jebadiah-4b-v1 well calibrated?
No calibration measurement for jebadiah-4b-v1 is recorded yet.
Is jebadiah-4b-v1 open weight, and can it run on a CPU?
Yes, jebadiah-4b-v1 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 (Hugging Face). Pricing is per 1M tokens (USD); confirm with the provider before production use.