OpenAI o3 API

OpenAI o3 is a frontier reasoning model with exceptional performance on complex scientific, math, and coding tasks.

Modalities: Text, ImageText
In / Out price: $2/$8per 1M
Context window: 200K tokens
Released: Apr 16, 2025

OpenAI o3 overview

OpenAI o3 by OpenAI

OpenAI o3 is a reasoning-centric language model engineered to deliver accurate, context-aware responses while fluidly combining browsing, multimodal analysis, and tool usage in real time.

OpenAI o3 specs

Model ID: openai/o3-2025-04-16
Developer: OpenAI
Context window: 200,000 tokens
Max output: 100,000 tokens
Input: Text, Image
Output: Text
API: OpenAI-compatible, base URL https://api.aimlapi.com/v1

СhatGPT o3 Description

OpenAI's o3 is an advanced AI reasoning model engineered for complex problem-solving and multi-step analysis. With simulated reasoning capabilities and integrated tool use, it delivers unprecedented performance in coding, mathematics, and scientific research.

Performance Benchmarks

OpenAI o3 is optimized for reasoning, coding, and scientific problem-solving across multiple domains.

Context Window: 200k context with multi-modal processing with step-by-step reasoning chains
Reasoning Architecture: Simulated reasoning with internal reflection and self-analysis
Performance Benchmarks: SWE-bench Verified: 69.1%, AIME 2025: 88.9%, GPQA Diamond: 83.3%

o3 Performance Metrics

The o3 model demonstrates strong performance across multiple domains, achieving 91.6% accuracy on AIME 2024 math competition, 88.9% on AIME 2025, 83.3% on PhD-level science questions (GPQA Diamond), 20.32% on expert-level questions across subjects (Humanity's Last Exam), and a 2706 ELO rating in coding competitions (Codeforces)

Key Capabilities

OpenAI o3 delivers advanced reasoning capabilities for complex analytical workflows.

Simulated Reasoning: Internal reflection and step-by-step logical analysis before generating responses
Visual Reasoning Integration: Processes images directly within reasoning chains for multi-modal problem solving
Advanced Coding: Excels in software engineering with 69.1% accuracy on real-world coding benchmarks
Mathematical Excellence: Superior performance on competition-level mathematics with 88.9% AIME accuracy
Scientific Research: PhD-level science question handling with 83.3% accuracy on GPQA Diamond
Tool Integration: Native access to web browsing, Python execution, and file operations for agentic problem-solving
Self-Fact Checking: Built-in verification capabilities to improve response accuracy

Optimal Use Cases

Complex Problem-Solving: Multi-step analytical tasks requiring deep reasoning and logical deduction
Software Engineering: Large-scale coding projects, debugging, and architectural decision-making
Mathematical Research: Competition-level mathematics and advanced computational problems
Scientific Analysis: Research-grade scientific reasoning and hypothesis generation
Business Consulting: Strategic analysis requiring multi-faceted evaluation and critical thinking
Creative Ideation: Novel hypothesis generation in technical and creative domains

Comparison with Other Models

Vs. OpenAI o1: 20% fewer major errors, superior coding performance (69.1% vs 48.9% SWE-bench), enhanced mathematical reasoning (88.9% vs 74.3% AIME)
Vs. GPT-4o: Specialized for reasoning tasks requiring deliberation time, integrated visual reasoning capabilities
Vs. Claude Opus 4: Competitive coding accuracy (69.1% vs 72.5% SWE-bench estimated), stronger mathematical performance, integrated tool use for agentic workflows

Use cases

Software development automation including large codebase management
Scientific and mathematical research assistance
Business intelligence and complex decision-making support
Development of intelligent agents capable of autonomous problem solving
Educational tools with stepwise reasoning and explanation generation

Capabilities

Function calling
Streaming
Structured outputs
Reasoning
Web search

OpenAI o3 benchmarks by Artificial Analysis

See how OpenAI o3 compares in generation speed and latency. Data retrieved Nov 3, 2026.

Speed#53 / 219163Output tokens / second Latency#194 / 2166.80Seconds to first token Humanity's Last Exam#143 / 45820.1%Score GPQA Diamond#119 / 43982.7%Score

Speed

Output tokens / second

Benchmark results

Compare 17 models available on AI/ML API across 5 independent benchmarks.

Humanity's Last Exam

Expert-written questions across academic subjects.

OpenAI o3 API pricing

OpenAI o3 pricing in US dollars
Usage	Price	Unit
Input	$2	1M tokens
Output	$8	1M tokens
Cached input	$0.5	1M tokens

Prices from the AI/ML API price list, checked on Nov 4, 2026.

OpenAI o3 vs other models: token prices

AI/ML API input and output rates for 5 models, per million tokens.

AI/ML API model prices in USD per million tokens
Model	Input	Output
OpenAI o3	$2	$8
GPT-6 Luna	$0.1	$0.5
GPT-6.1 Terra	$1	$5
GPT-6.1 Sol	$2	$10
GPT-6.1 Terra Pro	$2.12	$10.58

How to use the OpenAI o3 API

View docs

Build with your agent

Give your agent the model details, code and documentation.

1Get your API key

Create API key

export AIMLAPI_KEY="8f2b4c7d9a1e6b3f…"

2Make your first request

curl 'https://api.aimlapi.com/v1/chat/completions' \
  -H "Authorization: Bearer $AIMLAPI_KEY" \
  -H 'Content-Type: application/json' \
  -d '{
  "model": "openai/o3-2025-04-16",
  "messages": [
    {
      "role": "user",
      "content": "Write a haiku about the ocean."
    }
  ]
}'

import os
import requests

headers = {
    "Authorization": f"Bearer {os.environ['AIMLAPI_KEY']}",
    "Content-Type": "application/json",
}
payload = {
    "model": "openai/o3-2025-04-16",
    "messages": [
        {
            "role": "user",
            "content": "Write a haiku about the ocean."
        }
    ]
}

response = requests.post("https://api.aimlapi.com/v1/chat/completions", headers=headers, json=payload, timeout=120)
response.raise_for_status()
print(response.json())

const response = await fetch("https://api.aimlapi.com/v1/chat/completions", {
  method: "POST",
  headers: {
    Authorization: `Bearer ${process.env.AIMLAPI_KEY}`,
    "Content-Type": "application/json",
  },
  body: JSON.stringify({
    "model": "openai/o3-2025-04-16",
    "messages": [
      {
        "role": "user",
        "content": "Write a haiku about the ocean."
      }
    ]
  }),
});

if (!response.ok) throw new Error(`HTTP ${response.status}`);
console.log(await response.json());

All posts

OpenAI o3 FAQ

What is the context window of OpenAI o3?

OpenAI o3 has a 200,000 tokens context window and can return up to 100,000 tokens.

What inputs and outputs does OpenAI o3 support?

OpenAI o3 takes image, text as input and returns text.

What is the model ID for OpenAI o3?

Use openai/o3-2025-04-16 as the model id. Requests go to https://api.aimlapi.com/v1/chat/completions.

How much does OpenAI o3 cost?

OpenAI o3 costs $2 input and $8 output per 1M tokens on AI/ML API, billed as you go with no subscription.

Does OpenAI o3 support streaming?

Yes, OpenAI o3 can stream responses as they are generated.

Can OpenAI o3 read images?

Yes, OpenAI o3 accepts image input alongside text.

Who created OpenAI o3?

OpenAI o3 was built by OpenAI.

Start building with OpenAI o3

Get API key

OpenAI o3 API

openai/o3-2025-04-16

Build with your agent

Give your agent the model details, code and documentation.

1Get your API key

Create API key

export AIMLAPI_KEY="8f2b4c7d9a1e6b3f…"

2Make your first request

curl 'https://api.aimlapi.com/v1/chat/completions' \
  -H "Authorization: Bearer $AIMLAPI_KEY" \
  -H 'Content-Type: application/json' \
  -d '{
  "model": "openai/o3-2025-04-16",
  "messages": [
    {
      "role": "user",
      "content": "Write a haiku about the ocean."
    }
  ]
}'

import os
import requests

headers = {
    "Authorization": f"Bearer {os.environ['AIMLAPI_KEY']}",
    "Content-Type": "application/json",
}
payload = {
    "model": "openai/o3-2025-04-16",
    "messages": [
        {
            "role": "user",
            "content": "Write a haiku about the ocean."
        }
    ]
}

response = requests.post("https://api.aimlapi.com/v1/chat/completions", headers=headers, json=payload, timeout=120)
response.raise_for_status()
print(response.json())

const response = await fetch("https://api.aimlapi.com/v1/chat/completions", {
  method: "POST",
  headers: {
    Authorization: `Bearer ${process.env.AIMLAPI_KEY}`,
    "Content-Type": "application/json",
  },
  body: JSON.stringify({
    "model": "openai/o3-2025-04-16",
    "messages": [
      {
        "role": "user",
        "content": "Write a haiku about the ocean."
      }
    ]
  }),
});

if (!response.ok) throw new Error(`HTTP ${response.status}`);
console.log(await response.json());

Docs for AI agents Plain-text index · llms.txt API documentation API reference and examples

Talk to our sales team

Thanks for reaching out

OpenAI o3 API

OpenAI o3 overview

OpenAI o3 by OpenAI

OpenAI o3 specs

СhatGPT o3 Description

Performance Benchmarks

o3 Performance Metrics

Key Capabilities

Optimal Use Cases

Comparison with Other Models

Use cases

Capabilities

OpenAI o3 benchmarks by Artificial Analysis

Speed

Benchmark results

Humanity's Last Exam

GPQA Diamond

AA-LCR

IFBench

τ²-Bench Telecom

OpenAI o3 API pricing

OpenAI o3 vs other models: token prices

How to use the OpenAI o3 API

Build with your agent

1Get your API key

2Make your first request

Related articles

OpenAI o3 FAQ

Start building with OpenAI o3

OpenAI o3 API

Build with your agent

1Get your API key

2Make your first request

OpenAI o3 API

OpenAI o3 overview

OpenAI o3 by OpenAI

OpenAI o3 specs

СhatGPT o3 Description

Performance Benchmarks

o3 Performance Metrics

Key Capabilities

Optimal Use Cases

Comparison with Other Models

Use cases

Capabilities

Speed

Benchmark results

OpenAI o3 API pricing

OpenAI o3 vs other models: token prices

How to use the OpenAI o3 API

Build with your agent

1Get your API key

2Make your first request

Models similar to OpenAI o3

GPT-6.1 Terra

GPT-6.1 Terra Pro

GPT-6.1 Sol

Claude Sonnet 5.5

Claude Sonnet 5

Gemini 3.1 Pro

Related articles

OpenAI o3 FAQ

Start building with OpenAI o3