Skip to main content
POST
cURL

Authorizations

Authorization
string
header
required

Use your APIPod API key as a Bearer token in the Authorization header.

Body

application/json
model
string
required

Public APIPod model ID.

Allowed value: "gpt-image-2.6-sunburst-lite"
prompt
string
required

Used to describe the generated images. For GIF output, describe the subject, the motion and how the loop should close.

Maximum string length: 32000
image_urls
string[]

Publicly reachable source or reference image URLs. Not supported together with gif output.

Maximum array length: 16
mask_url
string

Inpaint mask image as a public URL or base64 data URI. Transparent pixels mark the region the model may edit; the rest of the image stays unchanged. Requires image_urls and cannot be combined with markup_url.

markup_url
string

Markup overlay image as a public URL or base64 data URI. A transparent PNG placed on top of the first reference image whose opaque strokes guide the model to the regions to edit; dimensions must match the first image after orientation. Requires image_urls and cannot be combined with mask_url.

output_format
enum<string>
default:png

Output format. gif is only available for text-to-image and does not support output_compression or transparent backgrounds; svg is not available on GPT Image 2.6 Lite.

Available options:
png,
jpeg,
webp,
gif
complexity
enum<string>
default:high

How much the model plans before rendering. low suits focused edits such as changing a background; high helps with dense text and complex layouts. Defaults to high and costs nothing extra.

Available options:
low,
medium,
high
quality
enum<string>
default:auto

Output fidelity tier. Must be at least medium when background is transparent.

Available options:
auto,
low,
medium,
high,
xhigh,
max
background
enum<string>
default:auto

Background mode. transparent returns an alpha-channel image and requires quality of medium or higher.

Available options:
auto,
opaque,
transparent
aspect_ratio
enum<string>
default:auto

Aspect ratio of the generated image.

Available options:
auto,
61,
123,
182,
184,
243,
245,
304,
556,
969,
1269
resolution
enum<string>
default:1K

Resolution tier of the generated images.

Available options:
1K,
2K,
4K
size
string

Optional escape hatch for exact pixel dimensions, as WIDTHxHEIGHT (for example 1024x1536). Each side must be a multiple of 16 and at most 3840. Takes precedence over resolution and aspect_ratio when both are sent. resolution + aspect_ratio already covers the 15 supported ratios at 1K/2K/4K, and works for GIF frames too, so size is only needed for pixel-exact or off-ratio sizes.

frame_count
integer
default:9

Number of GIF frames, 2-64. Only read when output_format is gif; defaults to 9.

Required range: 2 <= x <= 64
frame_duration_ms
integer
default:200

Duration of each GIF frame in milliseconds, 20-2000. Only read when output_format is gif; defaults to 200.

Required range: 20 <= x <= 2000
output_compression
integer

Compression level 0-100 for jpeg and webp output. Not supported by gif.

Required range: 0 <= x <= 100
moderation
enum<string>
default:low

Moderation strictness. APIPod defaults to low for GPT Image 2.6 to reduce false blocks; pass auto to restore OpenAI's default filtering.

Available options:
low,
auto

Response

200 - application/json

Task accepted

Standard response for task submission

code
integer
required
data
Task Submit Data · object
required

Data content for task submission

message
string
required