ZTC-Judge-4B
ZTC-Judge-4B by FINAL_Bench 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
- FINAL_Bench
- Parameters
- 5B
- 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
- —
- Jev-compatible request
- No
- Runs on CPU
- No
- License
- apache-2.0
- Hugging Face
- FINAL-Bench/ZTC-Judge-4B
Benchmarks
- DecisionBenchECE 0.180
- 0.428
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.
- Typed Decision Leaderboard (answer verifiers)Self-reportedAnswer-verification AUC (P(answer correct)) on 2,018 items, field-weighted — maintained by VIDRAFT (makers of ZTC)
- 0.636
Compare ZTC-Judge-4B with…
FAQ
How much does ZTC-Judge-4B cost?
ZTC-Judge-4B has no hosted price listed; its weights are open, so the cost is the compute you run it on.
What questions can ZTC-Judge-4B answer?
ZTC-Judge-4B answers choice (one of n options), score (ordered rubric), yes / no questions, returning a probability for each allowed answer.
Is ZTC-Judge-4B well calibrated?
No calibration measurement for ZTC-Judge-4B is recorded yet.
Is ZTC-Judge-4B open weight, and can it run on a CPU?
Yes, ZTC-Judge-4B 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 (DecisionBench, Hugging Face). Pricing is per 1M tokens (USD); confirm with the provider before production use.