OpenAI
o3 API
o3 is OpenAI's flagship reasoning model — it spends inference compute on internal chain-of-thought before answering. State-of-the-art on math, competitive coding, and scientific problems where step-by-step reasoning matters.
- Input
- $10.00
- Output
- $40.00
- Context
- 200K
- Vision
- Text only
/ 1M tokens
/ 1M tokens
tokens
Top use cases
- Math and competitive coding
- Scientific problem solving
- Hard agentic planning tasks
- Code reasoning and debugging
Use o3 in 30 seconds
ModelServer is OpenAI-compatible. Point your existing OpenAI SDK at modelserver.dev/v1 and set the model name to o3.
o3.py
from openai import OpenAI
client = OpenAI(
api_key="sk-modelserver-...",
base_url="https://modelserver.dev/v1",
)
response = client.chat.completions.create(
model="o3",
messages=[
{"role": "user", "content": "Hello, o3!"}
],
)
print(response.choices[0].message.content)Frequently asked questions
- How much does o3 cost?
- o3 is priced at $10.00 per 1M input tokens and $40.00 per 1M output tokens via ModelServer. ModelServer adds a flat 5.5% platform fee on top — no markups on individual tokens, no monthly minimum.
- What is the o3 context window?
- o3 supports a 200K token context window. You can put roughly 150,000 words in a single prompt.
- Is o3 OpenAI-compatible via ModelServer?
- Yes. Point your OpenAI SDK base_url to https://modelserver.dev/v1 and set model="o3". Existing OpenAI-SDK code works without modification.
- Who is o3 best for?
- Problems where slow but correct beats fast but hand-wavy.
- Does o3 support vision input?
- No. o3 is text-only. For multimodal use cases consider Claude Sonnet 4, GPT-4o, or Gemini 2.5 Pro.