Image generation (text-to-image) with Velqa.dev

Velqa offers image generation on the OpenAI-compatible API https://api.velqa.dev/v1/images/generations, with three Black Forest Labs models covering drafts, everyday use and high quality. Ideal for illustrations, mockups and content visuals.

Pricing and access

Like audio, images are billed per unit rather than per token, from your Recharge Boost balance (prepaid credit) — they are not included in any Starter/Dev/Pro plan.

• Access: any account with a positive Boost balance
• Per-request limits: 4 images maximum (n), size 1024×1024 maximum
• The exact cost is known before the call (n × price) — unlike audio, no pessimistic reservation is needed.

If the Boost balance is insufficient to cover n × price, the request is rejected (HTTP 402) before it ever reaches the model. A request with an out-of-range n or size is rejected (HTTP 400).

Example with curl

curl https://api.velqa.dev/v1/images/generations \
  -H "Authorization: Bearer $VELQA_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{"model":"flux-2-klein-4b","prompt":"a cat on a Moroccan rug","n":1,"size":"1024x1024"}'

Example with Python

from openai import OpenAI

client = OpenAI(base_url="https://api.velqa.dev/v1", api_key="VELQA_API_KEY")
res = client.images.generate(model="flux-2-klein-4b", prompt="a cat on a Moroccan rug", n=1, size="1024x1024")
print(res.data[0].b64_json[:40])

The response contains data[].b64_json (base64-encoded image — no URL, decode it client-side). The real cost (number of images generated × price/image) is debited from your Boost balance.