Instant Character

Instant Character

Playground

Instant Character transforms an uploaded character image and a text prompt into styled, action- and mood-driven character renders. Ready-to-use REST inference API, best performance, no coldstarts, affordable pricing.

Features

Instant Character — wavespeed-ai/instant-character

Instant Character is an image-guided character generation model that uses a reference image to keep a subject’s identity consistent while you place them into new scenes with a text prompt. Upload a character image, describe what the character is doing and where the scene takes place, and the model generates a new image that preserves the look while adapting pose, clothing details, and environment to match your prompt. It’s especially useful for character consistency across multiple creative iterations.

Key capabilities

• Reference-image guided character consistency (identity and overall look)
• Prompt-driven scene creation (action, environment, styling)
• Works well for generating multiple variations of the same character
• Custom output size (set width/height directly)
• Seed control for reproducible outputs

Use cases

• Create a consistent character for storyboards, comics, or visual novels
• Generate marketing/lifestyle scenes with the same model/character
• Outfit, pose, and background variations while keeping identity stable
• Rapid creative iteration by changing prompt or seed
• Building character packs for downstream video or design workflows

Pricing

Inputs

• prompt (required): what the character should do and the scene context
• image (required): reference character image

Parameters

• prompt: text instruction (subject action + scene + style)
• image: reference image (upload or URL)
• width / height: output size (e.g., 1280×720)
• seed: random seed (-1 for random; fixed for reproducible results)

Prompting guide

For best consistency, be explicit about “who stays the same” and “what changes”:

Template: Use the same character from the reference image. Place her in [scene]. She is [action]. Style: [photorealistic/cinematic/etc.]. Keep identity consistent.

Example prompts

• Use the same girl from the reference image. She is playing a guitar on a city street, candid street photography style, warm afternoon light, shallow depth of field, realistic details.
• Same character, standing in a rainy alley under neon lights, cinematic mood, wet reflections on the ground, subtle fog, 35mm lens look.
• Same character, sitting in a cozy café reading a book, soft window light, natural colors, calm atmosphere.

# Submit the task
curl --location --request POST "https://api.wavespeed.ai/api/v3/wavespeed-ai/instant-character" \
--header "Content-Type: application/json" \
--header "Authorization: Bearer ${WAVESPEED_API_KEY}" \
--data-raw '{
    "size": "1024*1024"
}'

# Get the result
curl --location --request GET "https://api.wavespeed.ai/api/v3/predictions/${requestId}/result" \
--header "Authorization: Bearer ${WAVESPEED_API_KEY}"