R
Relayby Ai5labs
← Back to all models

openai/gpt-4o

Aliases: 4o, gpt-4o-latest

$2.50
Input / 1M
$10.00
Output / 1M
128K
Context
110 t/s
Speed

Public benchmark scores

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

Quality index
71
MMLU
88.7
GPQA
53.6
HumanEval
90.2
MATH
76.6
SWE-bench
Arena Elo
1338

Sources: openai-research-page, lmsys-arena

Capabilities

toolsvisionjson_modestructured_outputstreaming
Input modalities: text, image, file

Use gpt-4o via Relay

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

YAML
# models.yaml
version: 1
models:
  4o:
    target: openai/gpt-4o
    credential: $env.OPENAI_API_KEY
Python
from relay import Hub

async with Hub.from_yaml("models.yaml") as hub:
    resp = await hub.chat(
        "4o",
        messages=[{"role": "user", "content": "Hello"}],
    )
    print(resp.text, resp.cost_usd)

pip install ai5labs-relay · full docs on GitHub

Compare with