Models / OpenAI/ o3 Mini High

o3 Mini High

OpenAI · released Feb 12, 2025

Input: text and documents. Output: text.InputOutput
Type
Proprietary
Input
$1.10
Output
$4.40
Cached
$0.55

List price · per 1M tokens · OpenAI at 200K context · source ↗

Our take

Written Aug 3, 2026

o3 Mini High is a hosted-only reasoning model from OpenAI that handles up to 200,000 tokens in a single request. It excels at coding and mathematical tasks on the Arena leaderboards we track, though its creative writing score lags well behind its own technical strengths.

Who should pick it

Choose this for coding work where its Arena Coding score is the headline strength, or for long-context tasks needing 200,000 tokens. It suits users already in the OpenAI ecosystem who want consistent pricing across both available providers. Skip it if you need strong creative writing, want to self-host, or are looking for price competition between hosts.

The case for it

  • Coding performance leads its own benchmark suite: its Arena Coding score sits 122.8 points above its Creative Writing score.
  • Mathematical reasoning is similarly strong relative to creative tasks, at 93.5 points higher on the same leaderboard.
  • Scores are stable across evaluation dates: Arena Coding varied by only 0.0109 points between two July and August 2026 measurements.

The case against it

  • Creative writing is the weakest area in its own profile, trailing its Maths score by 93.5 points and its Coding score by 122.8.
  • No price competition between its two providers: both charge identical rates, so there is no cheaper option.
  • Throughput is unverified on one of two providers; only OpenAI direct lists a speed figure.
00

How good is it?

IntelligencePuzzles, maths, exam questions

2 of 5

Arena Text (overall)96th of 143 · 1363.6via High

Arena Hard Prompts 88th of 143 via HighArena Maths 71st of 139 via High

CodingWriting and fixing code on its own

2.5 of 5

Arena Coding83rd of 143 · 1434.7via High

Arena Coding is the only board that has scored it for this.

AgenticPlanning, calling tools, staying on task

not measured

Nobody we watch has scored o3 Mini High for this. We would take the rating from Arena Agent (IPS).

WritingWe do not rate this

Scored, not ratedThe placings are on the right.

Two boards come close and neither tests writing: Arena Creative Writing asks people which of two replies they prefer, and LiveBench Language tests whether a model understood a passage. So we show where o3 Mini High placed and give it no mark out of five.

Arena Creative Writing 101st of 143 · 1311.8 via High
Also scored, on boards we give no mark for
Arena Instruction Following 89th of 143 via High

These tests check whether a model follows instructions — a precondition for all the work above, but not a measure of how well that work is done, which is why they get no rating.

Every published score for this model6 scoresEvery figure we hold, from 6 boards, with who ran it and a link to the source — including the boards no rating above is built on.
1434.7via Highindependentsource ↗
1311.8via Highindependentsource ↗
1401.3via Highindependentsource ↗
1366.5via Highindependentsource ↗
1405.3via Highindependentsource ↗
1363.6via Highindependentsource ↗
01

Or rent it from someone else

Cheapest published offer

Why this differs from the header. The strip above quotes OpenAI's own list price. This is the cheapest live offer at the widest standard context we hold, whoever is serving it — a reseller undercutting a lab is ordinary commerce, not an error.

per 1M tokens
$1.10 in / $4.40 out
Context served
200K
Throughput
Not measured
Current provider offers with price, context and prompt-privacy answers
ProviderIn / out per 1M tokensContextThroughputTrains on promptsLogs promptsZero retention
OpenRouter$1.10 / $4.40200Knot measuredUnknownUnknownUnknown
OpenAI$1.10 / $4.40200K31 tok/sNoYesunknown periodUnknown

Across the 2 listings we hold: 1 say they do not train on prompts, 0 say they do and 1 do not say. 0 appear in the zero-retention registry we check; the rest are unknown to us rather than confirmed either way.

What each host's API supports

From the parameter list each endpoint publishes. Streaming is omitted: nothing we hold reports it, for any model.

Supported
Not supported
Not published
host gave no parameter list
API features per host
ProviderTool callingJSON outputStrict schema
OpenRouter
OpenAI

Tool calling: 2 of 2 listings say yes. JSON output: 2 of 2 listings say yes. Strict schema: 2 of 2 listings say yes.

02

Models people weigh against o3 Mini High

03

When we formed this view

Dates behind this page

Aug 2, 2026BenchmarkScored 1434.7 via High on Arena Codingleaderboard
Aug 2, 2026BenchmarkScored 1311.8 via High on Arena Creative Writingleaderboard
Aug 2, 2026BenchmarkScored 1401.3 via High on Arena Hard Promptsleaderboard
Aug 2, 2026BenchmarkScored 1366.5 via High on Arena Instruction Followingleaderboard
Aug 2, 2026BenchmarkScored 1405.3 via High on Arena Mathsleaderboard
Aug 2, 2026BenchmarkScored 1363.6 via High on Arena Text (overall)leaderboard
Jul 26, 2026ListedListed on LLMapfirst indexed by our pipeline
Feb 12, 2025Announcedo3 Mini High announced by OpenAI

Prices last checked 9d ago

What we do not know about this model yet

  • Nothing we hold says whether an endpoint streams, so we do not show it either way.
  • 1 of 2 listings do not say whether they train on prompts.
04

Licence and identifiers

What the licence allowsWe hold no licence record for this model. Inside are the identifiers you need to pull it — its Hugging Face repo where we have one, our slug and a machine-readable card.

Licence

Commercial API terms. We hold no licence record for this model, so there is nothing to summarise here.

Identifiers

Modality record
text+file->text
Catalogue slug
openai-o3-mini-high

Machine-readable model card (omc.json) →

Something wrong on this page? Tell us