Image generation

Generate images

QuickSilver Pro speaks the OpenAI Images API. Point the OpenAI SDK at our base URL and call /v1/images/generations — the same key and the same balance you use for chat work for images too.

Models

  • flux.2-pro — flagship quality. Photoreal product shots, detailed scenes, accurate text rendering. $0.027 per image.
  • flux.1-schnell — fast, high-volume. Sub-second generation for previews, thumbnails, and batch jobs. $0.0025 per image.

Image models are billed per generated image (not per token). Spend comes out of the same prepaid balance as your chat usage — one key, one bill, no separate account to manage.

Python — OpenAI SDK

python
import os, base64
from openai import OpenAI

client = OpenAI(
    base_url="https://api.quicksilverpro.io/v1",
    api_key=os.environ["QSP_KEY"],
)

result = client.images.generate(
    model="flux.2-pro",          # or "flux.1-schnell" for fast/cheap
    prompt="a sleek silver sports car, studio product shot",
    n=1,
    size="1024x1024",
)

# Response matches OpenAI: data[0].b64_json holds the base64 image.
img = base64.b64decode(result.data[0].b64_json)
with open("out.jpg", "wb") as f:
    f.write(img)

curl

shell
curl https://api.quicksilverpro.io/v1/images/generations \
  -H "Authorization: Bearer $QSP_KEY" \
  -H "Content-Type: application/json" \
  -d '{
    "model": "flux.1-schnell",
    "prompt": "a yellow rubber duck on a white background",
    "n": 1,
    "size": "1024x1024"
  }'

The JSON response is OpenAI-shaped: data[0].b64_json is the base64-encoded image. Decode it and write the bytes to a file.

Notes

  • Text-to-image today — pass prompt, n, and size. The wire format matches the OpenAI Images API, so existing OpenAI-SDK image code works with only a base-URL and key change.
  • Same OpenAI-compatible surface as chat completions — no separate SDK, no separate billing relationship. Generate text and images against one key.