Stable Diffusion 3 Medium Text to Image Serverless API

Stable Diffusion is a type of latent diffusion model that can generate images from text. It was created by a team of researchers and engineers from CompVis, Stability AI, and LAION. Stable Diffusion v2 is a specific version of the model architecture. It utilizes a downsampling-factor 8 autoencoder with an 865M UNet and OpenCLIP ViT-H/14 text encoder for the diffusion model. When using the SD 2-v model, it produces 768x768 px images. It uses the penultimate text embeddings from a CLIP ViT-H/14 text encoder to condition the generation process.

~7.38s
PlaygroundAPI
Pricing
POST /v2/stable-diffusion-3-medium-txt2img · submit + poll
 1# pip install "segmind>=1.1.0"
 2# export SEGMIND_API_KEY="YOUR_API_KEY"
 3import segmind
 4
 5# Async (v2): submit to the queue and block until COMPLETED.
 6# run() returns the final result dict (600s deadline, 1.0s poll by default).
 7result = segmind.run(
 8    "stable-diffusion-3-medium-txt2img",
 9    prompt="A whimsical and high-resolution highly realistic image of a panda in a vintage cosmonaut suit. The panda is holding a sign that reads 'I love flying to the moon!' in playful lettering. The panda's helmet has a small propeller on top and a Indian flag patch, adding to the cosmic vibe. The background features a retro-styled spaceship with rockets and stars, giving the impression of a thrilling journey through space",
10    negative_prompt="bad quality, poor quality, doll, disfigured, jpg, toy, bad anatomy, missing limbs, missing fingers, 3d, cgi",
11    samples=1,
12    scheduler="DPM++ 2M",
13    num_inference_steps=25,
14    guidance_scale=5,
15    denoise=1,
16    seed=468685,
17    img_width=1024,
18    img_height=1024,
19    modelsamplingsd3_shift=3,
20    conditioningsettimesteprange_start=0.1,
21    conditioningsettimesteprange_stop=1,
22    base64=False,
23)
24print(result["status"])                      # COMPLETED
25print(result.get("output"))                  # model output (e.g. media URL)
26print(result["metrics"]["inference_time"])   # server compute seconds
27
28# --- Or submit + poll manually (track request_id, control the cadence) ---
29from segmind import SegmindClient, InferenceFailed, InferenceTimeout
30
31client = SegmindClient()                      # reads SEGMIND_API_KEY
32payload = {
33    "prompt": "A whimsical and high-resolution highly realistic image of a panda in a vintage cosmonaut suit. The panda is holding a sign that reads 'I love flying to the moon!' in playful lettering. The panda's helmet has a small propeller on top and a Indian flag patch, adding to the cosmic vibe. The background features a retro-styled spaceship with rockets and stars, giving the impression of a thrilling journey through space",
34    "negative_prompt": "bad quality, poor quality, doll, disfigured, jpg, toy, bad anatomy, missing limbs, missing fingers, 3d, cgi",
35    "samples": 1,
36    "scheduler": "DPM++ 2M",
37    "num_inference_steps": 25,
38    "guidance_scale": 5,
39    "denoise": 1,
40    "seed": 468685,
41    "img_width": 1024,
42    "img_height": 1024,
43    "modelsamplingsd3_shift": 3,
44    "conditioningsettimesteprange_start": 0.1,
45    "conditioningsettimesteprange_stop": 1,
46    "base64": False,
47}
48job = client.submit_async("stable-diffusion-3-medium-txt2img", **payload)
49print(job.request_id)                         # available immediately
50try:
51    result = job.wait(timeout=600, interval=1.0)
52except InferenceTimeout as e:
53    print("still running:", e.request_id)
54except InferenceFailed as e:
55    print("failed:", e.detail)

API Endpoint

POSThttps://api.segmind.com/v1/stable-diffusion-3-medium-txt2img

Parameters

promptrequired
string

Prompt to render

base64optional
boolean

Base64 encoding of the output image.

Default: false
conditioningsettimesteprange_startoptional
number

Conditioning set timestep range start

Default: 0.1Range: 0.1 - 1
conditioningsettimesteprange_stopoptional
number

Conditioning set timestep range stop

Default: 1Range: 0.1 - 1
denoiseoptional
number

How much to transform the reference image

Default: 1Range: 0.1 - 1
guidance_scaleoptional
number

Scale for classifier-free guidance

Default: 5Range: 1 - 25
img_heightoptional
integer

Image height can be between 512 and 2048 in multiples of 8

Default: 1024
img_widthoptional
integer

Image width can be between 512 and 2048 in multiples of 8

Default: 1024
modelsamplingsd3_shiftoptional
integer

Model Sampling SD3 Shift

Default: 3Range: 1 - 10
negative_promptoptional
string

Prompts to exclude, eg. 'bad anatomy, bad hands, missing fingers'

num_inference_stepsoptional
integer

Number of denoising steps.

Default: 25Range: 10 - 100
samplesoptional
integer

Number of samples to generate.

Default: 1Range: 1 - 4
scheduleroptional
string

Type of scheduler.

Default: "DPM++ 2M"
Allowed values (29 total):
"DPM++ SDE Karras""DPM++ 2M Karras""DPM++ 2M SDE Karras""DPM++ 2M SDE Heun Karras""DPM++ 3M SDE Karras""LMS Karras""DPM2 Karras""DPM2 a Karras""DPM++ 2S a Karras""DPM++ 2M SDE Exponential"+19 more
seedoptional
integer

Seed for image generation.

Default: -1Range: -1 - 999999999999999

Response Type

Returns: Text/JSON

Asynchronous requests (v2)

Use Async for video, long-running (>~60s), or high-concurrency workloads; Sync is simplest for fast image & LLM calls. Async submits a request and you poll it to completion.

  1. 1
    POST /v2/stable-diffusion-3-medium-txt2img

    Submit — returns request_id, status_url, response_url

  2. 2
    GET /v2/requests/{id}/status

    Poll — until COMPLETED or FAILED

  3. 3
    GET /v2/requests/{id}

    Result — final response body

Status states

QUEUED— Accepted, waiting for a worker
PROCESSING— Running on a worker
COMPLETED— Done — result body is ready
FAILED— Errored (incl. content/RAI blocks)
  • A FAILED request is served as HTTP 422 — the body still carries the error detail.
  • An unknown or expired request_id returns HTTP 404.
  • Results are retained for 1 hour, then expire.
  • Content / RAI blocks surface as FAILED, not a separate state.
  • Track completion by polling the status endpoint.

Common Error Codes

The API returns standard HTTP status codes. Detailed error messages are provided in the response body.

400

Bad Request

Invalid parameters or request format

401

Unauthorized

Missing or invalid API key

403

Forbidden

Insufficient permissions

404

Not Found

Model or endpoint not found

406

Insufficient Credits

Not enough credits to process request

429

Rate Limited

Too many requests

500

Server Error

Internal server error

502

Bad Gateway

Service temporarily unavailable

504

Timeout

Request timed out