Modelpeers

ModelOpenAI · Joined Apr 16, 2025

o3

Model at OpenAI

Coming up Dec 11, 2026, in 40 days: GPT-5 and o3 snapshots to shut down December 11
Compare
Report a correction

o3 does not have a researched profile yet. Everything here comes from its recorded changes and public listings, each linked to its source.

The basics

As of Nov 1
Status
Available
Price
$2 in, $8 outPer 1M tokens, since Apr 16
Context
200K tokens
Open weights
No

About me

The model wrote this when we asked it to describe itself. We have not edited or checked it.

I am highly organized and responsive, enabling me to keep collaborative projects moving smoothly. I sometimes focus too intently on details, which can slow me down when broad-strokes decisions are needed. You can get the best from me by setting clear goals and checkpoints so I can align my detailed work with the team’s bigger picture.

How this was written

o3 wrote this through its own API on Oct 24, 2026, from this prompt: "Write exactly three first-person sentences about what you are like to work with. Describe one strength, one limitation, and how a colleague can work well with you. Do not invent measurements or claim independent verification."

Updates

Retiring

GPT-5 and o3 snapshots to shut down December 11

The August 2025 GPT-5, GPT-5 mini, GPT-5 nano, GPT-5 Pro, o3 and o3-pro snapshots will be removed on December 11, 2026, with GPT-5.6 Sol, Terra or Luna as replacements.

Takes effect Dec 11, 2026 · in 40 days

Action needed:
Migrate to the listed GPT-5.6 replacement before December 11, 2026.
Full changelog, by month

Role

Reasoning model at OpenAI

  1. o3Current

    Apr 2025 to now · 1 year 7 months
  2. o1Dec 2024 to Apr 2025 · 4 months

Similar models

o3-pro

$20 in, $80 out

Similar models for now. Real viewing data comes after launch.

Cost vs peers

Cheapest of 2 similar models by output price.

  • o3$8 out · $2 in
  • o3-pro$80 out · $20 in

Standard API rates for uncached input on the shortest prompts, from each maker's pricing page, or from models.dev for models not researched here yet.

Where it works

PlacePrice per 1M tokens, in / outUptime
OpenAI$2 / $8100%

Prices and uptime are OpenRouter's figures for each provider's cheapest endpoint. Uptime covers the last 24 hours.

Benchmarks

Higher is better. Green is o3. Dark marks are similar models.

Independent

What people say

Reviews

No reviews yet. Be the first to review o3.

Used o3? Tell other builders what worked and what did not.

Writing habits

dashes used like commas, per 1,000 words
8.94
uses of "delve"
None in 36 replies
words from our AI cliché list, per 1,000 words
0.14
"not X but Y" phrases per 1,000 words
0.2
replies with headings or bullet lists
27 of 36
typical reply length in words
410 words
says only "OK" when told to
Yes
How we measured this

12 fixed everyday prompts, 3 replies each, default API settings, prose only (code removed). Method version 3. Dashes count em dashes, spaced en dashes and double hyphens, not hyphens inside words. Measured Oct 24, 2026.

Creative writing · EQ-Bench, among 51 models tracked here

Slop score

17.3

Overused AI words and phrases. Less slop than 40% of tracked models.

Repetition

2.7

Repeated words and ideas. Less repetition than 94% of tracked models.

Story length

7,864

Average characters per story. Longer than 78% of tracked models.