Stable Diffusion V3

Name: Stable Diffusion V3
Author: Stability AI

stability-ai/stable-diffusion-v3-medium/image-to-image

Stable Diffusion V3 Medium is Stability AI's intelligent image editing model. Transform, retouch, and reimagine existing images using text prompts - from background replacement to artistic style conversion.

Use in agent Get API key

Base price: $0.04 / run
Execution: async
Model type: image
Input fields: 8

Try the model

Playground

Open playground

Input

Prompt(string)

The prompt to generate an image from.

Image URL or upload(string)

Upload

PNG, JPEG, WebP, or GIF · 20 MiB maximum

The image URL to generate an image from.

This field is required.

Guidance scale (CFG)(number)

020

The CFG (Classifier Free Guidance) scale is a measure of how close you want the model to stick to your prompt when looking for a related image to show you. Range: 0 to 20.

Num Inference Steps(integer)

150

The number of inference steps to perform. Range: 1 to 50.

Strength(number)

0.011

The strength of the image-to-image transformation. Range: 0.01 to 1.

Enhance Prompt(boolean)

If set to true, prompt will be upsampled with more details.

Aspect Ratio(string)

The aspect ratio of the generated image. Allowed values: 21:9, 16:9, 3:2, 4:3, 5:4, 1:1, 4:5, 3:4, 2:3, 9:16.

Seed(integer)

The same seed and the same prompt given to the same version of Stable Diffusion will output the same image every time.

OutputReady

Your output will appear here

Complete the inputs, then click Run.

Specifications

Pricing

Base price: $0.04 / 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: Stability AI
Model ID: stability-ai/stable-diffusion-v3-medium/image-to-image
Execution: async
API: Unified Run API
Endpoint: /v1/run

API README

Stable Diffusion V3

Stable Diffusion V3 occupies a clearly bounded position in a image production pipeline under the route stability-ai/stable-diffusion-v3-medium/image-to-image. It is best evaluated as a named workflow endpoint rather than as a generic substitute for neighboring versions: planners can assign the route to a specific brief, record the chosen revision, and keep approval history tied to one predictable integration surface from intake through delivery.

A practical rollout of Stable Diffusion V3 should begin with source-readiness checks, ownership of the submitted material, an agreed budget, and a written definition of the final image to image deliverable. Before automation is enabled, the acceptance pass should verify that the returned asset opens correctly, fits the intended downstream toolchain, satisfies the project’s review criteria, and can be reproduced from the stored request and route identity stability-ai/stable-diffusion-v3-medium/image-to-image.

Highlights

Distinct output behavior. For Stable Diffusion V3, for Stable Diffusion V3, use an image URL as the starting visual In Stable Diffusion V3, this capability is applied through the exact stability-ai/stable-diffusion-v3-medium/image-to-image workflow with seed as a relevant request control; the seed control exposes the corresponding choice in this route; the seed control exposes the corresponding choice in this route.
Creative control. For Stable Diffusion V3, for Stable Diffusion V3, control the degree of transformation through strength In Stable Diffusion V3, this capability is applied through the exact stability-ai/stable-diffusion-v3-medium/image-to-image workflow with image as a relevant request control; the image control exposes the corresponding choice in this route; the image control exposes the corresponding choice in this route.
Workflow fit. For Stable Diffusion V3, for Stable Diffusion V3, set inference steps from 1 through 50 In Stable Diffusion V3, this capability is applied through the exact stability-ai/stable-diffusion-v3-medium/image-to-image workflow with prompt as a relevant request control; the prompt control exposes the corresponding choice in this route; the prompt control exposes the corresponding choice in this route.
Production detail. For Stable Diffusion V3, for Stable Diffusion V3, adjust CFG scale between 0 and 20 In Stable Diffusion V3, this capability is applied through the exact stability-ai/stable-diffusion-v3-medium/image-to-image workflow with strength as a relevant request control; the strength control exposes the corresponding choice in this route; the strength control exposes the corresponding choice in this route.

Pricing

Configuration	Price
Per request	$0.035

When to Use

Scenario	Why this model fits
Choose Stable Diffusion V3	Use it when the required deliverable is specifically the image-to-image result documented for `stability-ai/stable-diffusion-v3-medium/image-to-image`.
Match the source material	Select this route when your inputs naturally map to `seed`, `image`, `prompt`, `strength`.
Use its distinguishing capability	For Stable Diffusion V3, use an image URL as the starting visual In Stable Diffusion V3, this capability is applied through the exact stability-ai/stable-diffusion-v3-medium/image-to-image workflow with seed as a relevant request control; the seed control exposes the corresponding choice in this route
Plan repeatable production	For Stable Diffusion V3, control the degree of transformation through strength In Stable Diffusion V3, this capability is applied through the exact stability-ai/stable-diffusion-v3-medium/image-to-image workflow with image as a relevant request control; the image control exposes the corresponding choice in this route
Confirm cost and delivery	Use the pricing combinations below and the output contract above when budgeting or automating Stable Diffusion V3 jobs.

Prompt Guide

Build the request around the exact stability-ai/stable-diffusion-v3-medium/image-to-image schema. Start with required fields, then add only the controls needed for the intended output; keep URLs reachable and enum values exactly as shown in Technical Specs.

{
  "image": "https://static.sandbase.ai/examples/stability/stable-diffusion-v3-medium/image-to-image/input_image_0.png",
  "prompt": "cat wizard, gandalf, lord of the rings, detailed, fantasy, cute, adorable, Pixar, Disney, 8k",
  "seed": 0,
  "strength": 0.9
}

Technical Specs

Specification	Value
Model ID	`stability-ai/stable-diffusion-v3-medium/image-to-image`
Execution mode	`async`
Required inputs	`prompt`, `image`
Request fields	`seed` (integer, optional); `image` (string, required); `prompt` (string, required); `strength` (number, optional, min 0.01, max 1); `aspect_ratio` (string, optional, options: 21:9 / 16:9 / 3:2 / 4:3 / 5:4 / 1:1 / 4:5 / 3:4 / 2:3 / 9:16); `guidance_scale` (number, optional, min 0, max 20); `prompt_expansion` (boolean, optional); `num_inference_steps` (integer, optional, min 1, max 50)
Output	`url` (string); `content_type` (string)

Related Models

Start building

Send your first request

OpenAI-compatible endpoint with unified authentication and usage tracking.

Production API

Endpoint

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

Model ID

stability-ai/stable-diffusion-v3-medium/image-to-image

# Preserve Unicode and shell metacharacters in the request body.
result=$(printf '%s' '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' \
  | base64 --decode \
  | curl --fail-with-body --silent \
      -X POST "https://api.sandbase.ai/v1/run" \
      -H "Authorization: Bearer $SANDBASE_API_KEY" \
      -H "Content-Type: application/json" \
      --data-binary @-)
run_id=$(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 language models

Model	Context	Input / 1M	Output / 1M	Released
Stable Diffusion V3This model `Stability AI`	—	—	—	Jun 12, 2024
Stable Diffusion V3 `Stability AI`	—	—	—	Aug 20, 2024
ControlNet SDXL `Stability AI`	—	—	—	Feb 28, 2024
ControlNet SDXL `Stability AI`	—	—	—	Dec 1, 2023
Stable Diffusion XL `Stability AI`	—	—	—	Apr 16, 2024
Stable Diffusion XL `Stability AI`	—	—	—	Apr 16, 2024