OpenAI Decisions API
Decision model API in Developer tools
Developers recommend it as one fast, cheap API for choosing between options, far quicker than LLM calls. Critics say its confidence scores are unreliable.
Developers recommend it as one fast, cheap API for choosing between options, far quicker than LLM calls. Critics say its confidence scores are unreliable.
44 people gave an opinion
- recommend it
- 25
- criticize it
- 19
- approval share
- 57%
Recommended in 2 of 13 months
- November 2025: not recommended
- December 2025: not recommended
- January 2026: not recommended
- February 2026: not recommended
- March 2026: not recommended
- April 2026: not recommended
- May 2026: not recommended
- June 2026: not recommended
- July 2026: not recommended
- August 2026: not recommended
- September 2026: recommended
- October 2026: recommended
- November 2026: not recommended (month in progress)
Why it doesn’t carry the badge
- 25 people recommend itNeeds 10 or more, met.
- 1.3 recommenders for every criticNeeds 2 to 1 or better, not met.
- Recommended in 2 of 13 monthsNeeds 4 or more, not met.
The same three rules apply to every product, and the badge can’t be bought. How the badge works
What people say about it
The reasons people give, with the number of comments that give each one. Open a reason to read what they said.
Why people recommend it
Each mark is one comment
Much faster than LLM calls
“The main advantage is the speed (~2-300ms per input) which I really need for this purpose.”
Comment from 11 Oct 2026 in “Build your own decision model” “It's much faster and cheaper (an order of magnitude).”
Comment from 7 Oct 2026 in “Decisions API is in public beta” “decisions is 10x faster than responses API”
Comment from 7 Oct 2026 in “Decisions API is in public beta” - Comment from 7 Oct 2026 in “Decisions API is in public beta”
Cheap per token, no output fees
“it's still very cheap, and if you have existing enterprise OpenAI relationship it's a lot easier to onboard something like this”
Comment from 6 Oct 2026 in “Decisions API is in public beta” “With decisions models, you don't pay for output tokens.”
Comment from 7 Oct 2026 in “Decisions API is in public beta” “It's $0.10 per million tokens. Dirt cheap.”
Comment from 6 Oct 2026 in “Decisions API is in public beta” One API for several models
“I'd much rather integrate one API than deal with separate integrations, auth, billing, and quirks for every provider.”
Comment from 2 Oct 2026 in “Explore decision models in one place with Decisions API” “Unified API simplifies model comparison and integration.”
Comment from 30 Sep 2026 in “Decision API – compare and call decision models through one API”
What people warn about
Each mark is one comment
Confidence scores seem unreliable
“The probabilities I am seeing so far do not correspond with figures the business would find very agreeable”
Comment from 7 Oct 2026 in “Decisions API is in public beta” “Luna cosplaying a decisions model and pretending the confidence score isn’t just a hallucination”
Comment from 9 Oct 2026 in “Typesafe AI raises $870M at $7.5B” “it tends to return either 0 or 1 and doesn't return middle confidence levels very much at all”
Comment from 9 Oct 2026 in “Typesafe AI raises $870M at $7.5B” Biases and order effects like LLMs
“These models seem to have the same biases and limitations of LLMs minus speed.”
Comment from 7 Oct 2026 in “Decisions API is in public beta” “It's still a predictive model under the hood, so order will always matter.”
Comment from 7 Oct 2026 in “Decisions API is in public beta” Not better than LLM calls
“this new method is much, much nicer but doesn't open up brand new use cases from what I see”
Comment from 8 Oct 2026 in “Decisions API is in public beta” “It seems a bit silly since OpenAI LLMs already can output structured data.”
Comment from 29 Sep 2026 in “Dots: Always-on agents” “it was worse in all dimensions (price, speed, and quality of the result)”
Comment from 14 Oct 2026 in “We’re releasing Jev 1.14.0 along with experimental features for dreamers” - Comment from 9 Oct 2026 in “Typesafe AI raises $870M at $7.5B”
- Comment from 6 Oct 2026 in “Decisions API is in public beta”
Seen as marketing or rushed copy
“It's another Jev copy, like we've seen so many over the last few weeks.”
Comment from 29 Sep 2026 in “DevDay 2026 Recap” “labelling something that is a simplification of existing models as”
Comment from 7 Oct 2026 in “Decisions API is in public beta” “All about marketing that's why it's hard to root for Jev's success”
Comment from 7 Oct 2026 in “Decisions API is in public beta” - Comment from 7 Oct 2026 in “Decisions API is in public beta”
What people said
40 of the 48 comments that mention it, newest first. Each links to the original comment.
“it was worse in all dimensions (price, speed, and quality of the result)”
“The main advantage is the speed (~2-300ms per input) which I really need for this purpose.”
“it tends to return either 0 or 1 and doesn't return middle confidence levels very much at all”
“Luna cosplaying a decisions model and pretending the confidence score isn’t just a hallucination”
“Costs more, higher error rates, slower, and the answers are worse.”
“this new method is much, much nicer but doesn't open up brand new use cases from what I see”
“That's clever; why wax poetic when you can ask a question and just get back the multiple choice”
“The probabilities I am seeing so far do not correspond with figures the business would find very agreeable”
“It's still a predictive model under the hood, so order will always matter.”
“These models seem to have the same biases and limitations of LLMs minus speed.”
“Asking it to choose an outcome behaved differently from drawing randomly”
“Why 'risk' using this/Jev when a Luna LLM call will be smarter (in theory)?”
“OpenAI thinks their API is really worth more than 2x the price?”
“This is one of the most intuitive doc pages I’ve ever seen.”
“These decision models are gonna be everywhere.”
“4 out of 5 of those are all going to resolve Yes, with a far greater than 0.5 rate.”
“this is the biggest win, imo”
“impressive stuff! I gave it a bunch of examples”
“they are much faster and cheaper”
“All about marketing that's why it's hard to root for Jev's success”
“Clearly this was a rush job to respond to the competition.”
“It's far more cost-efficient and performant to use this type of model instead of an LLM.”
“labelling something that is a simplification of existing models as”
“It's much faster and cheaper (an order of magnitude).”
“isn't really anything new it just seems like a new API”
“Overpriced crap, local models are better than this”
“With decisions models, you don't pay for output tokens.”
“decisions is 10x faster than responses API”
“it's still very cheap, and if you have existing enterprise OpenAI relationship it's a lot easier to onboard something like this”
“Basically slower, more expensive and less capable than Jev”
“It's $0.10 per million tokens. Dirt cheap.”
“usually when OpenAI define an endpoint like that it ends up as a defecto standard for other providers.”
“Running a decision model is way easier and much cheaper.”
“it already supports image inputs”
“This article seems to be missing important points regarding how these models are intended to be used.”
“I'd much rather integrate one API than deal with separate integrations, auth, billing, and quirks for every provider.”
“Unified API simplifies model comparison and integration.”
“Decisions API is a validation for Jev and the entire space it created.”
“It's another Jev copy, like we've seen so many over the last few weeks.”
“It seems a bit silly since OpenAI LLMs already can output structured data.”
Counts come from 1,795,640 Hacker News comments, November 2025 to November 2026. How we count. Report a mistake.