Metask Lab's calibrated decision model in 16 languages. Qwen3.5-4B with a merged LoRA trained with the Nimble candidate-logit objective; one forward pass and a softmax over at most 26 answer-letter logits, with one temperature per question type.
meraGPT's hosted decision model (state-decider-1). Answers yes/no, choice and rubric-score questions over one state as calibrated distributions in a single pass, on the System One schema, so the typesafe-sdk works by changing its base URL.
Decides
choice, score, noul, classify, route
choice, score, noul, classify, route
Architecture
metask-jev
state-decider
Fine-tuned from
qwen/qwen3.5-4b
—
License
apache-2.0
proprietary
Availability
Open weights
Hosted API
Hosted by
—
meraGPT
Input price
—
$0.030/MTok
Decision accuracy
80.1%
76.8%
Calibration error
—
—
Valid action rate
—
—
Median latency
62.8 ms
—
p95 latency
—
—
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 metask-jev and state-decider-1?
metask-jev is from Metask Lab and state-decider-1 from meraGPT. metask-jev has open weights you can download and run; state-decider-1 is only available as a hosted API. Both answer choice, score, noul, classify and route questions. metask-jev is licensed apache-2.0; state-decider-1, proprietary.
Which is more accurate, metask-jev or state-decider-1?
They report on different suites — metask-jev 80.1% on JevBench v1.2 public set (231 items) at 4,096 tokens, maker's run, state-decider-1 76.8% on typed-decisions benchmark (400 cases, 2,000 decisions; teacher-ensemble references), single run — so the numbers do not rank them. Test both on your own labelled examples.
Which is cheaper, metask-jev or state-decider-1?
metask-jev: Free (open weights). state-decider-1: $0.03 / $0 per 1M. Open weights cost nothing per call beyond your own hardware.
Can I run metask-jev or state-decider-1 locally?
metask-jev yes — systemone pull metask-lab/metask-jev downloads its weights. The other is only served as a hosted API.
Evaluation suite
JevBench v1.2 public set (231 items) at 4,096 tokens, maker's run
typed-decisions benchmark (400 cases, 2,000 decisions; teacher-ensemble references), single run
