Stability AI SD3 image generation via MPP — text-to-image and image-to-image with multiple SD3.5 model variants.
What it does
This endpoint provides access to Stability AI's Stable Diffusion 3.5 family of image generation models through the Locus MPP (Micropayment Protocol), settling payments via Tempo on pathUSD. It supports both text-to-image and image-to-image modes across four model variants: sd3.5-large ($0.075/image), sd3.5-large-turbo ($0.046), sd3.5-medium ($0.040, default), and sd3.5-flash ($0.029). The endpoint accepts a JSON POST with a required text prompt and optional parameters including model selection, mode (text-to-image or image-to-image), negative prompt, aspect ratio, CFG scale, output format (jpeg/png/webp), and seed for reproducibility.
This is one endpoint within a broader Stability AI service suite available on the same Locus MPP gateway, which also includes Generate Ultra, Generate Core, image editing tools (erase, inpaint, outpaint, search-and-replace, search-and-recolor), background removal and replacement, upscaling (fast/conservative/creative), sketch-to-image, structure control, style guide, style transfer, 3D generation (Stable Fast 3D, Stable Point Aware 3D), and audio generation (text-to-audio, audio-to-audio, audio inpaint). The full OpenAPI spec is published at the gateway origin.
The endpoint is a POST-only route. The MPP probe returned 404 on HEAD/GET, which is expected since the endpoint only accepts POST requests with a payment challenge. The OpenAPI spec confirms the 402 Payment Required flow for this path. Pricing is model-dependent and ranges from $0.029 to $0.075 per generation as documented in the schema description field.
Capabilities
Use cases
- —Generating images from text prompts for content creation or marketing
- —Converting sketches or rough images into polished outputs via image-to-image mode
- —Rapid prototyping of visual concepts using the fast sd3.5-flash model
- —Producing high-quality hero images with sd3.5-large for editorial or commercial use
- —Automated image generation in agent workflows with per-call micropayments
Fit
Best for
- —AI agents needing pay-per-call image generation without API key management
- —Developers wanting Stable Diffusion 3.5 access via a standardized payment protocol
- —Workflows requiring multiple SD3.5 model tiers at different price/quality tradeoffs
- —Applications needing reproducible image generation with seed control
Not for
- —High-volume batch generation where subscription pricing would be cheaper
- —Real-time or sub-second latency requirements (generation takes seconds)
- —Users who need non-image Stability AI services (use sibling endpoints for audio, 3D, editing)
Quick start
curl -X POST https://stability-ai.mpp.paywithlocus.com/stability-ai/generate-sd3 \
-H "Content-Type: application/json" \
-H "Authorization: <MPP-TEMPO-TOKEN>" \
-d '{"prompt": "A futuristic cityscape at sunset, photorealistic", "model": "sd3.5-medium", "output_format": "png"}'Example
Request
{
"mode": "text-to-image",
"seed": 42,
"model": "sd3.5-medium",
"prompt": "A futuristic cityscape at sunset, photorealistic",
"cfg_scale": 7,
"aspect_ratio": "16:9",
"output_format": "png",
"negative_prompt": "blurry, low quality"
}Endpoint
Quality
Full OpenAPI schema with detailed request parameters and model-dependent pricing is available. However, the MPP probe did not capture a 402 challenge on HEAD/GET (expected for POST-only endpoints), no response schema or example response is documented, and crawled pages returned only 404 JSON errors. Pricing is documented in description fields but the amount is null (model-dependent), making exact payment challenge details unclear.
Warnings
- —MPP probe returned 404 on HEAD and GET — endpoint is POST-only; liveness not directly confirmed via probe
- —No response schema documented — output format and structure are unknown
- —Pricing amount is null in x-payment-info; model-dependent pricing ($0.029–$0.075) is described only in text
- —Currency address 0x20c000000000000000000000b9537d11c60e8b50 assumed to be pathUSD with 6 decimals based on other endpoints in the same spec
- —No example response available from probe or crawl data
Citations
- —SD3 endpoint supports models sd3.5-large ($0.075), sd3.5-large-turbo ($0.046), sd3.5-medium ($0.040), sd3.5-flash ($0.029)https://stability-ai.mpp.paywithlocus.com
- —Payment method is Tempo with intent chargehttps://stability-ai.mpp.paywithlocus.com
- —API reference available at platform.stability.aihttps://platform.stability.ai/docs/api-reference
- —LLM-specific docs at beta.paywithlocus.comhttps://beta.paywithlocus.com/mpp/stability-ai.md