The flagship of the vLLM Semantic Router team's Decision 1.0 family. Qwen3.5-9B with a shared candidate head returns a probability for every supplied answer to choice, yes/no and score questions, over a 16,384-token input.
The first System One model. Reads a state, answers typed Choice, Score and Noul questions in one call with calibrated probabilities, and generates no text. Closed weights, served by TypeSafe AI.
Decides
choice, score, noul, classify, route
choice, score, noul, classify, route
Architecture
decision
jev
Fine-tuned from
qwen/qwen3.5-9b
—
License
apache-2.0
proprietary
Availability
Open weights
Hosted API
Hosted by
—
TypeSafe AI
Input price
—
$0.042/MTok
Decision accuracy
77.4%
—
Calibration error
—
—
Valid action rate
—
—
Median latency
—
—
p95 latency
—
—
Evaluation suite
Figures are from each model’s manifest; accuracy and latency are what the publishers report, on their own suites and hardware. Add a third model.
Questions
What is the difference between decision and jev?
decision is from vLLM Semantic Router and jev from TypeSafe AI. decision has open weights you can download and run; jev is only available as a hosted API. Both answer choice, score, noul, classify and route questions. jev reads up to 32K tokens of state, against 16K tokens for decision. decision is licensed apache-2.0; jev, proprietary.
Which is more accurate, decision or jev?
Only decision publishes an accuracy figure (77.4% on vLLM-SR decision benchmark (54 tasks, 3,766 decisions, weighted)); jev does not, so there is no comparison to make without your own test.
Which is cheaper, decision or jev?
decision: Free (open weights). jev: $0.042 / $0 per 1M. Open weights cost nothing per call beyond your own hardware.
Can I run decision or jev locally?
decision yes — systemone pull vllm-semantic-router/decision downloads its weights. The other is only served as a hosted API.
