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.
OpenAI o3 is optimized for reasoning, coding, and scientific problem-solving across multiple domains.
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)
OpenAI o3 delivers advanced reasoning capabilities for complex analytical workflows.
Accessible via AI/ML API. Documentation: available here.
Measured on the AI/ML API gateway across live traffic. Figures are medians unless stated. Last updated 2026-10-12.
import requests r = requests.post( "https://api.aimlapi.com/v1/chat/completions", headers={"Authorization": "Bearer " + AIMLAPI_KEY}, json={ "model": "openai/o3-2025-04-16", "messages": [ { "role": "user", "content": "Hello!" } ] }, ) print(r.json())
const r = 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": "Hello!" } ] }), }); console.log(await r.json());
curl -X POST 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":"Hello!"}]}'
OpenAI-compatible — swap the base URL and it works with your existing SDK.
| Type | Price |
|---|---|
| Input | |
| Output | |
| Cached input |
| Benchmark | Score | What it measures | Source | Retrieved |
|---|---|---|---|---|
| SWE-bench Verified | 69.1% | Resolving verified real GitHub issues | Source | July 8, 2026 |
| GPQA Diamond | 83.3% | Google-proof graduate science questions (hardest subset) | Source | July 8, 2026 |
| AIME 2025 | 88.9% | Competition mathematics (AIME), 2025 | Source | July 8, 2026 |
| Intelligence | 20.2 | Composite score across standardised reasoning, knowledge and problem-solving evaluations, measured independently by Artificial Analysis | Source | September 22, 2026 |
| Math | 88.3 | Composite score across standardised mathematics evaluations, measured independently by Artificial Analysis | Source | September 22, 2026 |
| Model | Input | Output | Context | Best for |
|---|---|---|---|---|
OpenAI o3 This page | Reasoning + agents | |||
| Reasoning + agents | ||||
| Balanced coding + agents | ||||
| Long-context, multimodal & agentic workflows | ||||
| Reasoning + agents |
OpenAI o3 has a 200,000 tokens context window and can return up to 100,000 tokens.
OpenAI o3 takes image, text as input and returns text.
Use openai/o3-2025-04-16 as the model id. Requests go to https://api.aimlapi.com/v1/chat/completions.
OpenAI o3 is priced at input $2.6 / 1M tokens, output $10.4 / 1M tokens, cached input $0.65 / 1M tokens.
Yes, OpenAI o3 can stream responses as they are generated.
Yes, OpenAI o3 accepts image input alongside text.
OpenAI o3 was built by OpenAI.
