Skip to main content
Glama

Inpaint an image

ideogram_inpaint

Repaint the masked area of an image with Ideogram 3.0 variants: provide the source image and a mask where black marks the region to regenerate.

Instructions

Repaint the masked part of an image (black in the mask = repaint) with Ideogram 3.0 or its character / custom-model variants. Needs the source and the mask, each a local file, a public https URL or an Ideogram asset. Results are saved to the output directory; ideogram_quote prices a call first (nothing is generated).

Input Schema

NameRequiredDescriptionDefault
maskNoLocal mask image file or a public https URL, the same size as the source
seedNoRandom seed for a reproducible result
imageNoLocal image file to work on, or a public https URL of one
modelNoThe model (default ideogram-3); each takes the fields of its variant below
promptNoWhat to create or change, in natural language
wait_sNoSeconds to wait for the result (0-50, default 45). The job is accepted first; if it is still running when the wait ends, the result is its generation_id for ideogram_generation (never resubmitted, never billed twice).
privateNoKeep the result out of Ideogram's public feed; true unless you set false (enterprise accounts are always private)
num_imagesNoHow many images to make (each is billed)
style_typeNoVisual style family
style_codesNo8-character hexadecimal style codes
magic_promptNoauto, on or off: let Ideogram rewrite the prompt
style_presetNoA named style preset; not with style codes or style references
inline_imagesNotrue: each saved image is also returned as image content (base64; one over 3.75 MB, or past 10 MB in total, by path only), for a client without file access
rendering_speedNoturbo, default or quality: faster is cheaper
custom_model_uriNoA custom model you trained or were given access to
mask_asset_identifierNoAn Ideogram asset to use as the mask
image_asset_identifierNoAn Ideogram asset (an upload or an earlier result) instead of a local image
style_reference_imagesNoLocal image files (or public https URLs) whose style the result follows
character_reference_maskNoLocal grayscale mask (or a public https URL) of where the character is in its reference image
character_reference_imagesNoLocal image files (or public https URLs) of the character to keep consistent
enable_copyright_detectionNoRun Ideogram's copyright detection on the result (adds latency)
style_reference_collection_idNoA saved style collection to apply
style_reference_asset_identifiersNoIdeogram assets (uploads or earlier results) to use as style references
character_reference_asset_identifiersNoIdeogram assets to use as the character reference
style_reference_collection_version_idNoA version of the saved style collection

Schema Changelog

Changes observed during successful MCP inspections.

  1. Addedv2.5.0

TDQS

A3.9/5.0
Behavior3/5

Does the description disclose side effects, auth requirements, rate limits, or destructive behavior?

No annotations are provided, so the description carries the full burden. It usefully discloses the mask convention (black = repaint), accepted input sources, and where results are saved, and warns that ideogram_quote generates nothing. It omits key behaviors such as per-image billing, the async wait_s/generation_id handoff, and privacy defaults, which matter for a 25-parameter generation tool.

Agents need to know what a tool does to the world before calling it. Descriptions should go beyond structured annotations to explain consequences.

Conciseness5/5

Is the description appropriately sized, front-loaded, and free of redundancy?

Two dense sentences with no waste, front-loading the core action and mask semantics before the input requirements and the pricing alternative. Every clause earns its place.

Shorter descriptions cost fewer tokens and are easier for agents to parse. Every sentence should earn its place.

Completeness3/5

Given the tool's complexity, does the description cover enough for an agent to succeed on first attempt?

With no output schema and no annotations, the description must compensate for a complex tool with three model variants and 25 parameters. It covers the essential call semantics and result location but leaves async behavior, billing, and privacy handling to the schema, leaving the picture incomplete for an agent.

Complex tools with many parameters or behaviors need more documentation. Simple tools need less. This dimension scales expectations accordingly.

Parameters4/5

Does the description clarify parameter syntax, constraints, interactions, or defaults beyond what the schema provides?

Schema description coverage is 100%, so the baseline is 3. The description adds meaning beyond the schema by explaining the mask convention (black = repaint) and that image/mask may be a local file, https URL, or Ideogram asset, details not present in the field descriptions. It adds value but covers only a subset of the many parameters.

Input schemas describe structure but not intent. Descriptions should explain non-obvious parameter relationships and valid value ranges.

Purpose5/5

Does the description clearly state what the tool does and how it differs from similar tools?

States a specific verb and resource ('Repaint the masked part of an image') and pins down scope precisely, distinguishing it from siblings like ideogram_edit, ideogram_remove_object and ideogram_replace_background. An agent can tell exactly what this does without opening the schema.

Agents choose between tools based on descriptions. A clear purpose with a specific verb and resource helps agents select the right tool.

Usage Guidelines3/5

Does the description explain when to use this tool, when not to, or what alternatives exist?

Gives input prerequisites ('Needs the source and the mask') and routes pricing to ideogram_quote, which is useful context. However it never states when to choose inpaint over the many adjacent edit/background siblings, so alternative selection is left to inference.

Agents often have multiple tools that could apply. Explicit usage guidance like "use X instead of Y when Z" prevents misuse.