openai/o3-mini

$1.10

Input / 1M

$4.40

Output / 1M

200K

Context

95 t/s

Speed

Public benchmark scores

Sourced from each provider's published numbers. Verify before quoting.

Quality index

MMLU

86.5

GPQA

79.7

HumanEval

87.8

MATH

97.9

SWE-bench

—

Arena Elo

—

Sources: openai-research-page

Capabilities

toolsstructured_outputstreamingthinking

Input modalities: text, file

Use o3-mini via Relay

Configure the model alias in YAML, then call it from Python.

YAML

# models.yaml
version: 1
models:
  o3:
    target: openai/o3-mini
    credential: $env.OPENAI_API_KEY

Python

from relay import Hub

async with Hub.from_yaml("models.yaml") as hub:
    resp = await hub.chat(
        "o3",
        messages=[{"role": "user", "content": "Hello"}],
    )
    print(resp.text, resp.cost_usd)

pip install ai5labs-relay · full docs on GitHub

Compare with

vs anthropic/claude-opus-4-5 vs deepseek/deepseek-reasoner vs google/gemini-2.5-pro vs anthropic/claude-sonnet-4-5