ERNIE-Image Trainer

Name: Baidu ERNIE-Image Trainer
Price: 1.2 USD
Availability: InStock
Author: Baidu

baidu/ernie-image-trainer

Ernie Image Trainer by Baidu - advanced AI model for training. Delivers high-quality results with fast inference, suitable for both creative and production workflows.

Use in agent Get API key

Base price: $1.20USD / run
Execution: async
Model type: image
Input fields: 4

Try the model

Playground

Open playground

Input

Images Data Url(string)

URL to the input data zip archive. The zip should contain pairs of images and corresponding captions. The images should be named: ROOT.EXT. For example: 001.jpg The corresponding captions should be named: ROOT.txt. For example: 001.txt If no text file is provided for an image, the default_caption will be used.

Default Caption(string)

Default caption to use when caption files are missing. If None, missing captions will cause an error.

Learning Rate(number)

Learning rate.

Steps(integer)

1040000

Number of steps to train for Range: 10 to 40000.

OutputReady

Your output will appear here

Complete the inputs, then click Run.

Model card

Specifications

Pricing

Base price: $1.20 / run

Context & modalities

Input: Schema-defined
Output: image

Capabilities

Chat: Not supported
Vision: Not supported
Reasoning: Not supported
Structured output: Not supported
Function calling: Not supported
Audio input: Not supported

Access

Provider: Baidu
Model ID: baidu/ernie-image-trainer
Execution: async
Base URL: https://api.sandbase.ai
API: Unified Run API
Endpoint: /v1/run

API README

ERNIE-Image Trainer

baidu/ernie-image-trainer is a dedicated custom image-model training route in Baidu's ERNIE image family. It is designed around the named transformation, using the supplied text or media references to produce a coherent creative result rather than exposing a generic endpoint with loosely related controls.

The route combines its task-specific conditioning with the family's strengths in knowledge-aware image synthesis, Chinese-language understanding, typography, and detailed composition. This makes it useful for production workflows that need deliberate art direction, recognizable subjects, consistent scene logic, and a finished asset that can move directly into review or downstream editing.

Highlights

Subject adaptation. Learns a reusable visual concept from a focused set of training images.

Identity consistency. Encodes recurring subject features for later generation across new scenes and compositions.

Promptable reuse. Turns the trained concept into a controllable element for downstream image creation.

Style specialization. Supports adaptation toward a distinctive aesthetic or production-specific visual language.

Pricing

Billing unit	Price
Per request	$1.2

When to Use

✅ Good fit	❌ Consider alternatives
The model's named workflow matches the source material and intended output	A different input modality or model route is required
A managed asynchronous result is suitable for the production pipeline	A synchronous, interactive editor is essential
The documented controls cover the required duration, framing, or format	The project needs controls outside this endpoint's schema
Creative iteration benefits from a repeatable request structure	Exact deterministic pixels, frames, geometry, or samples are mandatory
A finished downloadable media asset is the desired deliverable	Editable source layers or a native project file are required

Prompt Guide

For generation, state the intended result first, then add the subject or source treatment, progression, style, and delivery constraints. Keep one creative variable per phrase, use the documented field names for controls, and change one setting at a time when comparing results.

{}

Technical Specs

Spec	Value
Model ID	`baidu/ernie-image-trainer`
Inputs	default_caption, images_data_url, learning_rate, steps
Required inputs	None
Output fields	content_type, url
Execution	Async (submit, then poll for result)

Related Models

baidu/ernie-image — Compare this concrete local family route.
baidu/ernie-image/turbo — Compare this concrete local family route.

Start building

Send your first request

OpenAI-compatible endpoint with unified authentication and usage tracking.

Production API

Unified Run API endpoint

https://api.sandbase.ai/v1/run

Model ID

baidu/ernie-image-trainer

# The quoted heredoc keeps Unicode and shell metacharacters unchanged.
result=$(curl --fail-with-body --silent \
  -X POST "https://api.sandbase.ai/v1/run" \
  -H "Authorization: Bearer $SANDBASE_API_KEY" \
  -H "Content-Type: application/json" \
  --data-binary @- <<'SANDBASE_JSON'
{
  "model": "baidu/ernie-image-trainer",
  "steps": 2000,
  "learning_rate": 0.0005
}
SANDBASE_JSON
)
run_id=$(printf '%s' "$result" | jq -r .id)
for attempt in $(seq 1 120); do
  status=$(printf '%s' "$result" | jq -r .status)
  case "$status" in completed|failed|timeout) break ;; esac
  sleep 2
  result=$(curl --fail-with-body --silent \
    -H "Authorization: Bearer $SANDBASE_API_KEY" \
    "https://api.sandbase.ai/v1/run/$run_id")
done
status=$(printf '%s' "$result" | jq -r .status)
[ "$status" = completed ] || { echo "Generation ended: $status" >&2; exit 1; }
printf '%s\n' "$result"

Choose your model

Compare models

All image & video models

Model	Type	Default price	Released
ERNIE-Image TrainerThis model `Baidu`	Image	$1.20 / run	Apr 15, 2026

Questions

FAQ

How do I call Baidu ERNIE-Image Trainer through SandBase?

Create a SandBase API key, then send requests with the model ID "baidu/ernie-image-trainer" to Unified Run API (/v1/run) at https://api.sandbase.ai. The request examples on this page show the exact payload.

Do I need a separate Baidu account?

No. One SandBase API key and balance gives you access to Baidu ERNIE-Image Trainer and the other models in the catalog; you do not need to sign up with Baidu separately.