nc://minimax-image

v0.1.0Pinned
ThomSharktools/nodecode-minimax-image: Draws pictures with MiniMax Image-01 from a prompt or a subject picture, saving the file under your images folder. - a rewrite of rriggs/hermes-plugin-minimax-image (MIT), credited to rriggs8d44aab· 10 Nov1 commit
test1 filecell.lisp31.0 KB

nodecode-minimax-image

MiniMax Image: pictures from MiniMax's Image-01 models, as the minimax_image tool and /minimax-image. A Nodecode rewrite of rriggs/hermes-plugin-minimax-image by rriggs.

The original is an image backend for another agent, plugged into its one image_generate tool. Here the same wire is a cell of its own: one tool, one command, and the pictures written under a folder you name.

Surface What it does
minimax_image One picture from a prompt: subject, setting, style, light, framing. The answer names the file, its size in bytes and its pixel size.
minimax_image with images A subject reference: up to 4 source pictures - local paths, https URLs or data URIs - and a prompt that says how the subject should look in the new picture.
/minimax-image The same from your side of the keyboard: /minimax-image a red fox in snow, with --landscape, --square, --portrait, --model and --image PATH in front. Typed alone it asks for the prompt.

Aspect picks the ratio MiniMax takes: landscape 16:9, square 1:1, portrait 9:16. image-01 draws from a prompt; image-01-live keeps a subject from a source picture, and a call that names one upgrades the model on its own. MiniMax reads 1500 characters of a prompt, and its answer links expire within a day, so the cell downloads the picture as soon as it is drawn.

Get it

Tell the agent "add the minimax-image cell", or run nodecode add nc://minimax-image.

Set it up

You need a MiniMax API key from https://platform.minimax.io (interface keys). Then either:

  • put it in the environment variable MINIMAX_API_KEY, which is what the cell reads unless the section names another variable, or
  • name a file holding it in the section's key_file.

The section minimax_image in ~/.nodecode/config.jsonc is optional; config.example.jsonc shows it:

{
  "minimax_image": {
    "api": "https://api.minimax.io",
    "key_env": "MINIMAX_API_KEY",
    "model": "image-01",
    "output_dir": "~/.nodecode/images"
  }
}

api is the host the key was made on, model is image-01 or image-01-live, and output_dir is where pictures land (~/.nodecode/images by default). With no key set the cell still starts: the tool and the command answer with the setting that names one, and nothing is dialled.

Use it

Ask for a picture - "draw a red fox in snow, wide shot" - and the model calls minimax_image. The answer carries the file's path under output_dir, its size and its pixel size, and the model says where it is.

From the keyboard: /minimax-image a red fox in snow, /minimax-image --portrait the same fox, at night --image ~/pictures/fox.png, /minimax-image --model image-01 a heron.

From eval, the same call:

(minimax-image:generate-image "a red fox in snow" :aspect "landscape")

A failed call says the HTTP status or MiniMax's own base_resp code, its message, and the setting that usually fixes it; the key itself never appears in an answer.

What the original does that this does not

Nothing of its request shape: the same endpoint, the same two models and their upgrade rule, the same three ratios, the same 1500-character prompt, the same 10 MB inline cap for a source image, the same base_resp error shape, and the same "download it now, the link expires" cache. It is not a provider in another agent's catalog here; it is a cell with a tool of its own. See NOTICE.

Add nc://minimax-image

  1. 1

    Install Nodecode

    Once per machine.

    $ curl -fsSL https://nodecode.ai/install.sh | sh
  2. 2

    Add the cell

    Pinned at 8d44aab.

    $ nodecode add nc://minimax-image

    Or tell the agent add the minimax image cell.