OpenAI's decision endpoint, announced at DevDay 2026. It focuses GPT-6 Luna on developer-defined questions that have a finite set of predefined answers, with text or image context, for classification, routing and an agent's next action.
A 144M-parameter System One model on the multilingual mmBERT-small encoder. Takes a state, a question and 2 to 20 options and returns one decision with probabilities for choice, score and yes/no questions, on a CPU.
Decides
choice, classify, route
choice, score, noul, classify, route
Architecture
gpt-6-luna
julia
Fine-tuned from
—
jhu-clsp/mmbert-small
License
proprietary
apache-2.0
Availability
Hosted API
Open weights
Hosted by
OpenAI
—
Input price
—
—
Decision accuracy
—
73.2%
Calibration error
—
—
Valid action rate
—
—
Median latency
—
—
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 decisions-api and julia-1?
decisions-api is from OpenAI and julia-1 from Supersonic Labs. decisions-api is only available as a hosted API; julia-1 has open weights you can download and run. Both answer choice, classify and route questions. Only julia-1 answers score and noul. decisions-api is licensed proprietary; julia-1, apache-2.0.
Which is more accurate, decisions-api or julia-1?
Only julia-1 publishes an accuracy figure (73.2% on typed-decisions test set (400 cases, 2,000 questions)); decisions-api does not, so there is no comparison to make without your own test.
Which is cheaper, decisions-api or julia-1?
decisions-api: Hosted, price not published. julia-1: Free (open weights). Open weights cost nothing per call beyond your own hardware.
Can I run decisions-api or julia-1 locally?
julia-1 yes — systemone pull supersonic-labs/julia-1 downloads its weights. The other is only served as a hosted API.
Evaluation suite
—
typed-decisions test set (400 cases, 2,000 questions)
