tasksource Decider nano v4 (150M)
tasksource Decider nano v4 (150M) by tasksource is a decision model: it answers typed questions with a probability for each allowed answer instead of generating text. It scores 17.6 on the Jev Decision Index (rank 127).
Pricing
No hosted price is listed. The weights are open: the cost is your own compute.
Specifications
- Provider
- tasksource
- Parameters
- 149M
- Released
- Sep 2026
- Open weights
- Yes
- Question types
- Choice (one of N options), Score (ordered rubric), Yes / no
- Mechanism
- Cross-encoder (NLI)
- Backbone
- cross-encoder/ettin-reranker-150m-v1
- Jev-compatible request
- Yes
- Runs on CPU
- Yes
- License
- apache-2.0
- Hugging Face
- tasksource/tasksource-decider-nano
Benchmarks
- Jev Decision IndexECE 0.069
- 17.6
Jev Decision Index: public 30.9, ECE 0.069, Brier 0.532. 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 0.3 kit (release-v3), self-run
- 30.9
Speed & latency
Compare tasksource Decider nano v4 (150M) with…
FAQ
How much does tasksource Decider nano v4 (150M) cost?
tasksource Decider nano v4 (150M) has no hosted price listed; its weights are open, so the cost is the compute you run it on.
What questions can tasksource Decider nano v4 (150M) answer?
tasksource Decider nano v4 (150M) answers choice (one of n options), score (ordered rubric), yes / no questions, returning a probability for each allowed answer.
Is tasksource Decider nano v4 (150M) well calibrated?
On the Jev Decision Index its expected calibration error (ECE) is 0.069; below about 0.05, a stated 80% confidence is right roughly 80% of the time.
Is tasksource Decider nano v4 (150M) open weight, and can it run on a CPU?
Yes, tasksource Decider nano v4 (150M) 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, Hugging Face). Pricing is per 1M tokens (USD); confirm with the provider before production use.