simple-jev · Qwen3.8-27B (featherless)
simple-jev · Qwen3.8-27B (featherless) by featherless-ai is a decision model: it answers typed questions with a probability for each allowed answer instead of generating text. It scores 51.8 on the Jev Decision Index (rank 40).
Pricing
No hosted price is listed. The weights are open: the cost is your own compute.
Specifications
- Provider
- featherless-ai
- Parameters
- 28B
- 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/Qwen3.8-27B
- Jev-compatible request
- No
- Runs on CPU
- No
- License
- —
Benchmarks
- Jev Decision IndexECE 0.113
- 51.8
- DecisionBenchECE 0.218
- 0.471
Jev Decision Index: public 56.0, ECE 0.113, Brier 0.400. Lower ECE means its probabilities can be taken at face value.
Speed & latency
Compare simple-jev · Qwen3.8-27B (featherless) with…
FAQ
How much does simple-jev · Qwen3.8-27B (featherless) cost?
simple-jev · Qwen3.8-27B (featherless) has no hosted price listed; its weights are open, so the cost is the compute you run it on.
What questions can simple-jev · Qwen3.8-27B (featherless) answer?
simple-jev · Qwen3.8-27B (featherless) answers choice (one of n options), score (ordered rubric), yes / no questions, returning a probability for each allowed answer.
Is simple-jev · Qwen3.8-27B (featherless) well calibrated?
On the Jev Decision Index its expected calibration error (ECE) is 0.113; below about 0.05, a stated 80% confidence is right roughly 80% of the time.
Is simple-jev · Qwen3.8-27B (featherless) open weight, and can it run on a CPU?
Yes, simple-jev · Qwen3.8-27B (featherless) 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, Benchmark Heaven JevBench). Pricing is per 1M tokens (USD); confirm with the provider before production use.