MODEL ID baidu:ernie-image@0

live

ERNIE-Image

by Baidu April 15, 2026

ERNIE-Image is Baidu's 8B text-to-image model built on a single-stream Diffusion Transformer architecture. It is designed for strong prompt adherence, reliable text rendering, and structured visual generation, making it well suited to posters, comics, storyboards, multi-panel layouts, and other workflows where content accuracy and composition matter as much as aesthetics. The standard model emphasizes stronger general-purpose capability and instruction fidelity, typically running at around 50 inference steps.

API Options

Platform-level options for task execution and delivery.

taskType string required value: imageInference: Identifier for the type of task being performed

taskUUID string required UUID v4: UUID v4 identifier for tracking tasks and matching async responses. Must be unique per task.

outputType string default: URL: Image output type.

Allowed values 3 values

outputFormat string default: JPG

Specifies the file format of the generated output. The available values depend on the task type and the specific model's capabilities.

`JPG`: Best for photorealistic images with smaller file sizes (no transparency).
`PNG`: Lossless compression, supports high quality and transparency (alpha channel).
`WEBP`: Modern format providing superior compression and transparency support.

**Transparency**: If you are using features like background removal or LayerDiffuse that require transparency, you must select a format that supports an alpha channel (e.g., `PNG`, `WEBP`, `TIFF`). `JPG` does not support transparency.

Allowed values 3 values

outputQuality integer min: 20 max: 99 default: 95: Compression quality of the output. Higher values preserve quality but increase file size.

webhookURL string URI

Specifies a webhook URL where JSON responses will be sent via HTTP POST when generation tasks complete. For batch requests with multiple results, each completed item triggers a separate webhook call as it becomes available.

Learn more 1 resource

Webhooks
PLATFORM

deliveryMethod string default: sync

Determines how the API delivers task results.

Allowed values 2 values

: Returns complete results directly in the API response.
: Returns an immediate acknowledgment with the task UUID. Poll for results using getResponse.

Learn more 1 resource

Task Polling
PLATFORM

uploadEndpoint string URI

Specifies a URL where the generated content will be automatically uploaded using the HTTP PUT method. The raw binary data of the media file is sent directly as the request body. For secure uploads to cloud storage, use presigned URLs that include temporary authentication credentials.

Common use cases:

Cloud storage: Upload directly to S3 buckets, Google Cloud Storage, or Azure Blob Storage using presigned URLs.
CDN integration: Upload to content delivery networks for immediate distribution.

// S3 presigned URL for secure upload
https://your-bucket.s3.amazonaws.com/generated/content.mp4?X-Amz-Signature=abc123&X-Amz-Expires=3600

// Google Cloud Storage presigned URL
https://storage.googleapis.com/your-bucket/content.jpg?X-Goog-Signature=xyz789

// Custom storage endpoint
https://storage.example.com/uploads/generated-image.jpg

The content data will be sent as the request body to the specified URL when generation is complete.

safety object

Content safety checking configuration for image generation.

Properties 1 property

safety » checkContent checkContent boolean default: false: Enable or disable content safety checking.

ttl integer min: 60: Time-to-live (TTL) in seconds for generated content. Only applies when outputType is URL.

includeCost boolean default: false: Include task cost in the response.

numberResults integer min: 1 max: 20 default: 1: Number of results to generate. Each result uses a different seed, producing variations of the same parameters.

acceleration string default: medium: Optimization level.

Allowed values 4 values

acceleratorOptions object

Advanced caching mechanisms to speed up generation.

Properties 12 properties

acceleratorOptions » cacheEndStep cacheEndStep integer min: 1: Absolute step number to end caching. Must be greater than cacheStartStep and less than or equal to steps.

acceleratorOptions » cacheEndStepPercentage cacheEndStepPercentage integer min: 1 max: 100: Percentage of steps to end caching. Alternative to cacheEndStep. Must be greater than cacheStartStepPercentage.

acceleratorOptions » cacheMaxConsecutiveSteps cacheMaxConsecutiveSteps integer min: 1 max: 5 default: 3: Limits the maximum number of consecutive steps that can use cached computations before forcing a fresh computation.

acceleratorOptions » cacheStartStep cacheStartStep integer min: 0: Absolute step number to start caching. Must be less than cacheEndStep.

acceleratorOptions » cacheStartStepPercentage cacheStartStepPercentage integer min: 0 max: 99: Percentage of steps to start caching. Alternative to cacheStartStep. Must be less than cacheEndStepPercentage.

acceleratorOptions » fbCache fbCache boolean default: false: First Block Cache (FBCache) acceleration. Reuses feature block computations across steps.

acceleratorOptions » fbCacheThreshold fbCacheThreshold float min: 0 max: 1 step: 0.01 default: 0.25: Controls the sensitivity threshold for determining when to reuse cached computations. Lower values reuse more aggressively.

acceleratorOptions » teaCache teaCache boolean default: false: TeaCache acceleration for transformer-based models. Estimates step differences to skip redundant computations.

acceleratorOptions » teaCacheDistance teaCacheDistance float min: 0 max: 1 step: 0.01 default: 0.5: Controls the aggressiveness of the TeaCache feature. Lower values prioritize quality, higher values prioritize speed.

acceleratorOptions » dbCache dbCache boolean default: false: DB Cache (CacheDiT) acceleration. Caches and reuses intermediate transformer block outputs to skip redundant computations.

acceleratorOptions » dbCacheThreshold dbCacheThreshold float min: 0 max: 1 step: 0.01 default: 0.25: Controls the sensitivity threshold for DB Cache. Lower values reuse cached blocks more aggressively, higher values prioritize quality.

acceleratorOptions » dbCacheSkipInterval dbCacheSkipInterval integer min: 1 default: 5: Controls how many steps to skip between cache refreshes. Higher values skip more steps for faster generation at the cost of quality.

Core Parameters

Primary parameters that define the task output.

model string required value: baidu:ernie-image@0: Identifier of the model to use for generation.

positivePrompt string required: Text prompt describing elements to include in the generated output.

width integer max: 2048 step: 16 default: 1024: Width of the generated media in pixels.

height integer max: 2048 step: 16 default: 1024: Height of the generated media in pixels.

seed integer min: 0 max: 9223372036854776000: Random seed for reproducible generation. When not provided, a random seed is generated in the unsigned 32-bit range.

steps integer min: 15 max: 50 default: 30: Total number of denoising steps. Higher values generally produce more detailed results but take longer.

CFGScale float min: 0 max: 20 step: 0.01 default: 4: Guidance scale representing how closely the output will resemble the prompt. Higher values produce results more aligned with the prompt.

Settings

Technical parameters to fine-tune the inference process. These must be nested inside the settings object.

settings » promptEnhance promptEnhance object

Automatic enhancement and expansion of the input prompt.

Properties 3 properties

settings » promptEnhance » enabled enabled boolean default: false: Automatic enhancement and expansion of the input prompt.

settings » promptEnhance » temperature temperature float min: 0 max: 5 step: 0.01 default: 1.2: Controls randomness in generation. Lower values produce more deterministic outputs, higher values increase variation and creativity.

settings » promptEnhance » topP topP float min: 0 max: 1 step: 0.01 default: 0.95: Nucleus sampling parameter that controls diversity by limiting the probability mass. Lower values make outputs more focused, higher values increase diversity.