kev-0.5b
kev-0.5b by Jared Palmer 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
- Jared Palmer
- Parameters
- 500M
- 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/Qwen2.5-0.5B
- Jev-compatible request
- Yes
- Runs on CPU
- Yes
- License
- apache-2.0
- Hugging Face
- jaredpalmer/kev-0.5b
Benchmarks
- Jev Decision Indexedition 0.1 (earlier edition)
- 5.2
- DecisionBenchECE 0.160
- 0.437
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.
- held-out split of the six training sources (1,350 questions)Self-reported
- 0.799
Compare kev-0.5b with…
FAQ
How much does kev-0.5b cost?
kev-0.5b has no hosted price listed; its weights are open, so the cost is the compute you run it on.
What questions can kev-0.5b answer?
kev-0.5b answers choice (one of n options), score (ordered rubric), yes / no questions, returning a probability for each allowed answer.
Is kev-0.5b well calibrated?
No calibration measurement for kev-0.5b is recorded yet.
Is kev-0.5b open weight, and can it run on a CPU?
Yes, kev-0.5b 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, DecisionBench, Benchmark Heaven JevBench, Hugging Face). Pricing is per 1M tokens (USD); confirm with the provider before production use.