GPT Image 2.5 Fast

GPT Image 2.5 Fast is built for generating images from a text prompt quickly, and it is the variant to reach for when you are exploring rather than finishing. It carries five detail levels where GPT Image 2 had three, so there is a genuine draft tier underneath the quality you would ship, and the two extra levels sit above it for a final frame. Billing follows the tokens it actually renders, so a draft settles at well under a cent while the highest detail level costs more, and you pay the measured amount rather than a flat rate. The listed price is one standard square image at the high level. It follows detailed prompts closely and renders readable text inside images.

Constraints & limits

The gateway auto-clamps requests to the model's limits where possible.

Quick start

This sample hits the image generation endpoint. Change base_url and your key, then run it.

from openai import OpenAI

client = OpenAI(
    api_key="tl-xxxxxxxxxxxxxxxxxxxxxxxx",
    base_url="https://api.thalam.ai/v1",
)

response = client.images.generate(
    model="openai/gpt-image-2.5-fast",
    prompt="A futuristic city skyline at sunset.",
    size="1024x1024",
    n=1,
)

print(response.data[0].url)