Ming-Image-0.1-Design on Sume: not in the catalog, use these

Sume does not serve Ming-Image-0.1-Design. Its image catalog lists named models (GPT Image 2.5, Nano Banana, Seedream, Flux, Ideogram, Recraft and others), and Ming is not one of them. For poster and infographic prompts, send a catalog id such as openai/gpt-image-2.5.

Ming facts below are from its Hugging Face model card; Sume facts are from the Image API docs. Both read 2026-10-01.

What is Ming-Image-0.1-Design?

The model card describes a 6B text-to-image model aimed at UI layouts, infographics, posters and other text-rich designs. It is released under the MIT license in BF16 weights, and it can produce transparent-background (RGBA) output when the prompt starts with the RGBA phrases the card documents.

The card recommends 2048x2048 output (1024 for speed), 12 steps, CFG 1.0, and one CUDA GPU with 80 GiB of VRAM. That is self-hosting guidance: you run the weights yourself.

Can I call it through the Sume API?

No. A model value outside the catalog returns 404 model_not_found, and GET /v1/images/models is the list of ids Sume accepts. Sume's docs describe no way to add a model of your own.

What do I send for a poster instead?

Pick from the catalog and read each model's supported_parameters before pinning a size. This request asks GPT Image 2.5 for a 4:3 poster at 1536x2048 pixels (both edges multiples of 16, 3,145,728 pixels, our arithmetic).

curl -X POST https://api.sume.com/v1/images \
  -H "Authorization: Bearer $SUME_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{
    "model": "openai/gpt-image-2.5",
    "prompt": "Event poster, bold headline 'AUTUMN SALE' at the top, three product tiles below, flat colors",
    "image_size": "1536x2048",
    "output_format": "png"
  }'

How do I compare before I switch?

Run the same prompt against two or three catalog ids and compare the results yourself; do not assume a benchmark transfers. Sume's text rendering comparison lists the models to try first.

How do Ming and the Sume catalog differ in what you get?

Facts from the sections above.