Enterprise DNAEnterprise DNA
OpenAI logo

Models · Reasoning

o3

by OpenAI

o3 is a well-rounded and powerful model across domains. It sets a new standard for math, science, coding, and visual reasoning tasks.

Visit OpenAI ↗

Our verdict

Not reviewed yet

We have not reviewed o3 yet

No one at Enterprise DNA has checked this listing against its sources. The description comes from public listings gathered automatically. Any fact below shows where it came from and when it was last checked.

How we review records

From the listing

About o3

o3 is a well-rounded and powerful model across domains. It sets a new standard for math, science, coding, and visual reasoning tasks. It also excels at technical writing and instruction-following.

Best for

Hard reasoning and verification steps in agent pipelines

Use cases

  • plan-then-act agents
  • math/logic checks
  • code verification

Enterprise DNA angle

Omni can stage o3 as a judge or planner while cheaper models do the volume work.

Notes

o3 is listed in the EDNA Models directory from the OpenRouter catalogue.

Pros

  • Strong deliberate reasoning for hard problems
  • Good fit for planning and verification steps

Cons

  • Latency and cost higher than flash/mini models
  • Not ideal for high-volume cheap inference

Key facts

What we have checked

Input price
$2.00 per million tokens
Checked OpenRouter models API
Output price
$8.00 per million tokens
Checked OpenRouter models API
Latest version
openai/o3-2025-04-16
Checked OpenRouter models API
Time to first token (ms)
7,490
Checked Artificial Analysis
Output speed (tokens a second)
152.1
Checked Artificial Analysis
Tokens served on OpenRouter in a day
110M (for 10 Nov 2026)
Checked OpenRouter model activity (748 of 748 series)
Tool calls on OpenRouter in a day
1.3K (for 10 Nov 2026)
Checked OpenRouter model activity (748 of 748 series)
Artificial Analysis Intelligence Index
20.2
Checked Artificial Analysis via OpenRouter
LMArena text rating (Elo)
1,410
Checked LMArena (CC-BY-4.0), published 2026-11-09
Gpqa diamond
81.8% (for 16 Apr 2025)
Checked Epoch AI (CC BY 4.0)
Swe bench verified
62.3% (for 16 Apr 2025)
Checked Epoch AI (CC BY 4.0)
Frontiermath tiers 1 3 v2 private
33.3% (for 16 Apr 2025)
Checked Epoch AI (CC BY 4.0)

Each value shows when we last confirmed it and where it came from. Values we have not confirmed are left out, never guessed.

API access

How to call it

Call via OpenRouter with model id `openai/o3`.

  • openrouter · openai/o3 · Docs

OpenRouter id: openai/o3

Price

What o3 costs, over time

Input price

$2.00 to $2.00

US dollars per million input tokens. 3 Sept 2026 to 10 Nov 2026, 44 days.

Source: OpenRouter models API. Checked .

Output price

$8.00 to $8.00

US dollars per million output tokens. 3 Sept 2026 to 10 Nov 2026, 44 days.

Source: OpenRouter models API. Checked .

Usage

How much o3 is used

Tokens served a day

46.9M to 110M

tokens a day on OpenRouter. 19 Sept 2026 to 10 Nov 2026, 53 readings.

Source: OpenRouter model activity (748 of 748 series). Checked .

Scores

How o3 scores

Each score comes from the named source and shows when we last confirmed it. A flag means the score needs care.

ScoreValueSourceChecked
Artificial Analysis Intelligence Index20.2indexAs of 26 Oct 2026Artificial Analysis via OpenRouter
LMArena text rating (Elo)1,410EloAs of 10 Nov 2026LMArena (CC-BY-4.0), published 2026-11-09
LMArena vision rating (Elo)1,214EloAs of 10 Nov 2026LMArena (CC-BY-4.0), published 2026-11-09
Design Arena rating (Elo)1,200EloAs of 10 Nov 2026Design Arena via OpenRouter
Aider polyglot (best-across-versions)81.3percentAs of 16 Apr 2025Epoch AI (CC BY 4.0)
Arc agi (best-across-versions)60.8percentAs of 16 Apr 2025Epoch AI (CC BY 4.0)
Arc agi 2 (best-across-versions)6.5percentAs of 16 Apr 2025Epoch AI (CC BY 4.0)
Cadeval (best-across-versions)74percentAs of 16 Apr 2025Epoch AI (CC BY 4.0)
Chess puzzles (best-across-versions)37percentAs of 16 Apr 2025Epoch AI (CC BY 4.0)
Cl bench (best-across-versions)17.8percentAs of 16 Apr 2025Epoch AI (CC BY 4.0)
Deepresearch bench (best-across-versions)45.2percentAs of 16 Apr 2025Epoch AI (CC BY 4.0)
Dtbench (best-across-versions)84.8percentAs of 16 Apr 2025Epoch AI (CC BY 4.0)
Show 20 more scores
ScoreValueSourceChecked
Fiction livebench (best-across-versions)88.9percentAs of 16 Apr 2025Epoch AI (CC BY 4.0)
Frontiermath 2025 02 28 private (best-across-versions)18.7percentAs of 16 Apr 2025Epoch AI (CC BY 4.0)
Frontiermath tier 4 2025 07 01 private (best-across-versions)2.1percentAs of 16 Apr 2025Epoch AI (CC BY 4.0)
Frontiermath tiers 1 3 v2 private (best-across-versions)33.3percentAs of 16 Apr 2025Epoch AI (CC BY 4.0)
Gdpval (best-across-versions)30.8percentAs of 16 Apr 2025Epoch AI (CC BY 4.0)
Geobench (best-across-versions)74percentAs of 16 Apr 2025Epoch AI (CC BY 4.0)
Gpqa diamond (best-across-versions)81.8percentAs of 16 Apr 2025Epoch AI (CC BY 4.0)
Gso bench (best-across-versions)8.8percentAs of 16 Apr 2025Epoch AI (CC BY 4.0)
Lech mazur writing (best-across-versions)83.9percentAs of 16 Apr 2025Epoch AI (CC BY 4.0)
Lmca (best-across-versions)39.7percentAs of 16 Apr 2025Epoch AI (CC BY 4.0)
Math level 5 (best-across-versions)97.8percentAs of 16 Apr 2025Epoch AI (CC BY 4.0)
Metr time horizons (best-across-versions)65.4percentAs of 16 Apr 2025Epoch AI (CC BY 4.0)
Mystery game puzzles (best-across-versions)29percentAs of 16 Apr 2025Epoch AI (CC BY 4.0)
Osworld (best-across-versions)23percentAs of 16 Apr 2025Epoch AI (CC BY 4.0)
Otis mock aime 2024 2025 (best-across-versions)Saturated84.4percentAs of 16 Apr 2025Epoch AI (CC BY 4.0)
Simplebench (best-across-versions)53.1percentAs of 16 Apr 2025Epoch AI (CC BY 4.0)
Simpleqa verified (best-across-versions)49.4percentAs of 16 Apr 2025Epoch AI (CC BY 4.0)
Swe bench verified (best-across-versions)62.3percentAs of 16 Apr 2025Epoch AI (CC BY 4.0)
Vpct (best-across-versions)52percentAs of 16 Apr 2025Epoch AI (CC BY 4.0)
Weirdml (best-across-versions)52.4percentAs of 16 Apr 2025Epoch AI (CC BY 4.0)
Saturated
Top models score close to the maximum, so this test no longer separates them.

Sources

Where this comes from

A source shows what was published when we checked it. It does not certify security, performance or a working installation. Read our method.

Changelog

What changed

  1. Deprecated

    OpenRouter lists the route openai/o3 to expire on 11 Dec 2026.

    Source: OpenRouter models API

USAGE / OPENROUTER

How much o3 is used

Tokens processed through OpenRouter each day. It shows what developers actually run, not every company's private traffic.

Tokens, last 7 days
755M ▲4.2%
Requests, last 7 days
226K
Tool calls, last 7 days
31.4K
People's vote
1410
Average cost per request
84.1¢
Price per million tokens
$2 in · $8 out
0125M250M19 Sept: 46.9M tokens, 10.3K requests20 Sept: 59.2M tokens, 9.4K requests21 Sept: 65.2M tokens, 15.9K requests22 Sept: 114M tokens, 19K requests23 Sept: 78.4M tokens, 14.1K requests24 Sept: 80.8M tokens, 11.5K requests25 Sept: 68.5M tokens, 9.6K requests26 Sept: 183M tokens, 12.7K requests27 Sept: 124M tokens, 14.8K requests28 Sept: 74.1M tokens, 13.8K requests29 Sept: 116M tokens, 17K requests30 Sept: 133M tokens, 36.2K requests1 Oct: 122M tokens, 20K requests2 Oct: 70.9M tokens, 15.8K requests3 Oct: 54.7M tokens, 13.7K requests4 Oct: 48.5M tokens, 17.5K requests5 Oct: 75M tokens, 24K requests6 Oct: 80.1M tokens, 24.9K requests7 Oct: 132M tokens, 28.6K requests8 Oct: 135M tokens, 29.7K requests9 Oct: 88.1M tokens, 25.4K requests10 Oct: 216M tokens, 33.3K requests11 Oct: 58.4M tokens, 20.2K requests12 Oct: 90.9M tokens, 23.8K requests13 Oct: 94.5M tokens, 26.1K requests14 Oct: 65.3M tokens, 22.1K requests15 Oct: 87M tokens, 30.1K requests16 Oct: 82.5M tokens, 30.1K requests17 Oct: 45.9M tokens, 21K requests18 Oct: 61.7M tokens, 23.3K requests19 Oct: 98.4M tokens, 28K requests20 Oct: 61.5M tokens, 21.3K requests21 Oct: 74.6M tokens, 24.9K requests22 Oct: 123M tokens, 32.5K requests23 Oct: 82.6M tokens, 20.8K requests24 Oct: 146M tokens, 41.8K requests25 Oct: 69.9M tokens, 21.1K requests26 Oct: 169M tokens, 35.9K requests27 Oct: 96.3M tokens, 24.6K requests28 Oct: 125M tokens, 22.7K requests29 Oct: 93M tokens, 27.7K requests30 Oct: 131M tokens, 27.7K requests31 Oct: 57.2M tokens, 22.1K requests1 Nov: 114M tokens, 35.8K requests2 Nov: 118M tokens, 30.2K requests3 Nov: 87.3M tokens, 23.7K requests4 Nov: 83.4M tokens, 26K requests5 Nov: 162M tokens, 33K requests6 Nov: 135M tokens, 30.5K requests7 Nov: 68.7M tokens, 32.6K requests8 Nov: 121M tokens, 43.3K requests9 Nov: 75.1M tokens, 28.5K requests10 Nov: 110M tokens, 31.7K requests19 Sept3 Oct17 Oct31 Oct10 NovSee where it ranks on the leaderboard →

MODEL FIELD GUIDE / SOURCE EVIDENCE

What you can do with o3.

Listed specifications; no EDNA task evaluation has been run. Sources checked 2026-11-11. Hosting routes and app features can differ.

DeveloperOpenAI
Exact model identifieropenai/o3
Inputimage, text, file
Outputtext
Tool calling
Listed support
Structured output
Listed support
Reasoning controls
Listed support
Availability
available
Price basis
token
Licence verification
Check the developer licence; the model name is insufficient
Context length
200K tokens
Knowledge cutoff
30 Jun 2024
Developer release date
Not established
First seen in the catalogue
16 Apr 2025
Listed retirement date
11 Dec 2026
Inspect the sources and their scope
catalogue source ↗

openai/o3: listed route capabilities, availability and prices; provider endpoints can differ.

Checked 2026-11-11
openrouter.ai ↗

Source for reasoning controls, first seen, tool calling, context length, retirement date, input, knowledge cutoff, output, structured output.

Checked 2026-11-11
How to evaluate a model for your work →

FROM SPECIFICATIONS TO YOUR WORK

Is this the right model for your job?

Start with a realistic sample. These workflows match the model’s listed capabilities, not a verified result on your task.

Write a clear business documentWorkflow & starter prompt ↗Analyse a spreadsheetWorkflow & starter prompt ↗Research a companyWorkflow & starter prompt ↗Build an appWorkflow & starter prompt ↗Extract document dataWorkflow & starter prompt ↗Run a reliable agentWorkflow & starter prompt ↗

What would it cost?

Pick a common job and how many times a month you run it. Token sizes are our assumption for a typical run, shown under the result.

Working out the cost...

Token prices only. Per-request, image, tool, storage and subscription fees are not included. One token is about three quarters of an English word. Real counts vary by model and language. How tokens are counted ↗

Check the source

Price date: 10 Nov 2026, from OpenRouter models API. Prices are USD API token rates. Your app subscription, uploads, browsing and tools are separate.

Model source & availability ↗

Understand the evidence

Listed capabilities and third-party benchmark scores do not establish Enterprise DNA task-test results. Validate accuracy and total cost on your own sample.

Design Arena: 1200 (2026-11-09) ↗
Compare this model with alternatives →
Free 27-page guide

Get the free Developer’s Field Guide

A 27-page field guide to the AI coding workflow with Claude. Claude Code, MCP servers, the prompt patterns that work, and what to delegate. Free.

Enter your work email. We send it straight over, plus a few short notes worth knowing. Unsubscribe any time.

No spam. Unsubscribe any time.

← Back to ModelsSubmit your own entry →