Bosun v3.1 1.7B
Bosun v3.1 1.7B by Hanno Labs is a decision model: it answers typed questions with a probability for each allowed answer instead of generating text. It scores 15.8 on the Jev Decision Index (rank 130).
Pricing
No hosted price is listed. The weights are open: the cost is your own compute.
Specifications
- Provider
- Hanno Labs
- Parameters
- 2B
- 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/Qwen3-1.7B-Base
- Jev-compatible request
- Yes
- Runs on CPU
- No
- License
- apache-2.0
- Hugging Face
- Hanno-Labs/bosun-v3.1-1.7b
Benchmarks
- Jev Decision IndexECE 0.130
- 15.8
- DecisionBenchECE 0.050
- 0.849
Jev Decision Index: public 20.6, ECE 0.130, Brier 0.614. 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.
- DecisionBenchSelf-reported
- 0.849
Speed & latency
Compare Bosun v3.1 1.7B with…
FAQ
How much does Bosun v3.1 1.7B cost?
Bosun v3.1 1.7B has no hosted price listed; its weights are open, so the cost is the compute you run it on.
What questions can Bosun v3.1 1.7B answer?
Bosun v3.1 1.7B answers choice (one of n options), score (ordered rubric), yes / no questions, returning a probability for each allowed answer.
Is Bosun v3.1 1.7B well calibrated?
On the Jev Decision Index its expected calibration error (ECE) is 0.130; below about 0.05, a stated 80% confidence is right roughly 80% of the time.
Is Bosun v3.1 1.7B open weight, and can it run on a CPU?
Yes, Bosun v3.1 1.7B 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 (Jev Decision Index, DecisionBench, Enterprise Decision Arena, Hugging Face). Pricing is per 1M tokens (USD); confirm with the provider before production use.