Microsoft AI: MAI-Image-2.6 Flash

Origin

microsoft/mai-image-2.6-flash

MAI-Image-2.6 Flash is the lower-latency, lower-cost member of the [MAI-Image-2.6](/microsoft/mai-image-2.6) family from Microsoft AI, built for latency-sensitive, high-throughput production image generation and editing at comparable quality to the precision tier....

Pricing

Output · 22.8 MA / 1M tokens

Billing follows the provider's own reported cost. The unit prices here are for reference; a model marked "billed at actual usage" has no reliable pre-call price, so your usage page is the record of what you paid.

Capabilities

VisionImage gen

Call it

Image models are served on POST /v1/images/generations. Pictures come back as inline base64 (data[].b64_json), never as URLs — the upstream provides none, so response_format: "url" is answered with a 400. Up to 3 images per request; streaming is not supported in this release.

curl https://macdecloud.com/v1/images/generations \
  -H "Authorization: Bearer dcld-sk-..." \
  -H "Content-Type: application/json" \
  -d '{"model":"microsoft/mai-image-2.6-flash","prompt":"a red panda reading a book","n":1,"resolution":"2K"}'

Python (OpenAI SDK)

Decode the returned base64 and write it to a file. This image-only client sets max_retries=0 to prevent automatic resubmission on failure. Keep the request ID and check the generation result and bill before deciding whether to retry.

import base64
from openai import OpenAI

client = OpenAI(base_url="https://macdecloud.com/v1", api_key="dcld-sk-...", max_retries=0)
resp = client.images.generate(model="microsoft/mai-image-2.6-flash", prompt="a red panda reading a book", n=1)
with open("image.png", "wb") as f:
    f.write(base64.b64decode(resp.data[0].b64_json))