← Back
Decisions API

Decisions API

software

1 creator use or recommend it

▲ 1 active in the last 90 days

1 recommend 3 mention
Last discussed Oct 4, 2026

OpenAI's API for making extremely fast decisions, enabling one decision roughly every 200 milliseconds.

What creators say

Dan Shipper
recommends
Used for: Fast structured outputs for LLM-as-judge and classifiers Praised: About 150ms responses, multimodal (accepts images) vs: Faster than Jev on some workloads

"It's really freaking fast. You can get responses in like 150 milliseconds. In our testing internally at Every, we did find that it was faster than Jev on some workloads, and crucially it's multimodal, so you can feed it images. So if you're a deve..."

More from Dan Shipper →
TS
Tibo Sottiaux
mentioned
Praised: Faster, supports visual inputs vs: Different shape from the Responses API

"For example, Decisions API came together very quickly. We realized we have a really good model with Luna. We can just do some constrained sampling and have it shipped as a different shape to the Responses API, and when we do that it's faster."

More from Tibo Sottiaux →
+1 more mention from Tibo

"We announced the Decisions API — we further tuned the constraints so that you can make decisions extremely fast, so you can make one decision rough..."

Riley Brown
mentioned
Praised: Fast classification vs: Unlike traditional LLMs, it has no output tokens

"The Decisions API is meant to make really fast intelligent decisions and it's only trained to select from possible options. So it would just select whether it should be sent to sales or billing or technical support."

More from Riley Brown →
Greg Isenberg
mentioned
Praised: Classifying content, routing requests, choosing agent next actions vs: OpenAI's answer to Jev; unclear whether it will be better or worse

"Developers supply context using text or images and they get back answers they can use to classify content, route requests, or choose an agent's next action. So it's in limited preview today. Excited to play around with it."

More from Greg Isenberg →

About this data

  • ✓48,000+ mentions extracted from real podcast transcripts
  • ✓Every quote links to the source episode
  • ✓Sponsored mentions are always labeled
  • ✓Critical takes are shown, never filtered out
How Followed works →
Get an email when a new creator recommends it

Follow Decisions API

Free email updates