How much does Stable Diffusion 3.5 Large (Stability) cost via Railwail?

Per-call: €0.07. No monthly minimum, no subscription. Start with €5 free credits.

What is the context window of Stable Diffusion 3.5 Large (Stability)?

Stable Diffusion 3.5 Large (Stability) supports a unknown context window — enough for typical AI workloads.

How fast is Stable Diffusion 3.5 Large (Stability)?

Latency depends on prompt length and load — typically 200ms to 2s for short prompts. We measure p50/p95 in real-time on /rankings.

Is Stable Diffusion 3.5 Large (Stability) better than Flux 1.1 Pro Ultra?

It depends on your use case. Stable Diffusion 3.5 Large (Stability) (Custom) and Flux 1.1 Pro Ultra (Black Forest Labs) are both strong choices in image generation. Compare them side-by-side at /compare/sd-3-5-large-stability-vs-flux-pro-ultra.

Stable Diffusion 3.5 Large (Stability)

Name: Stable Diffusion 3.5 Large (Stability)
Brand: Custom
SKU: sd-3-5-large-stability
Price: 0.065 EUR
Availability: InStock

Custom

Image Generation

Stability AI's 8B-parameter flagship SD3.5 model. Strong prompt adherence and aesthetic quality.

Generate with Stable Diffusion 3.5 Large (Stability)

Describe what you want and pick a size — the image renders inline.

Size

Result appears here

TL;DR·Last updated May 16, 2026

Stable Diffusion 3.5 Large (Stability) is image generation AI model from Custom, priced at €0.000 per 1M input tokens with a unknown context window.

Try Stable Diffusion 3.5 Large (Stability)

Prompt

Aspect Ratio

Quality

Direct API access coming soon

Pricing

Price per Generation

Per generation€0.07

API Integration

Use our OpenAI-compatible API to integrate Stable Diffusion 3.5 Large (Stability) into your application.

Install

npm install railwail

JavaScript / TypeScript

import railwail from "railwail";

const rw = railwail("YOUR_API_KEY");

const images = await rw.run("sd-3-5-large-stability", "A beautiful sunset over Tokyo");
console.log(images[0].url);

// Or use the image() method for full control
const res = await rw.image("sd-3-5-large-stability", "A cat in space", {
  size: "1024x1024",
  n: 1,
});
console.log(res.data[0].url);

Specifications

Price

€0.07

Developer

Custom

Deep dive — Stability AI's Stable Diffusion 3.5 Large (Stability)

About Stability AI

Founded 2019 · London, UK

Stability AI was founded in 2019 by Emad Mostaque (CEO until March 2024, succeeded by Prem Akkaraju) and is headquartered in London. The company famously sponsored and released Stable Diffusion 1.x and 2.x (2022) in partnership with CompVis (LMU Munich, Rombach et al.) and Runway, and then SDXL (2023), SD3 (2024) and SD 3.5 (Oct 2024). After the original team behind Stable Diffusion left in 2024 to found Black Forest Labs, Stability AI was restructured and recapitalised under new investors led by Sean Parker and Sean 'Diddy' Combs' backers. The SD 3.5 family — Large (8.1B), Large Turbo and Medium (2.5B) — was released in October 2024 under a permissive Stability AI Community License allowing free commercial use up to $1M revenue.

Visit Stability AI →

Architecture

Multimodal Diffusion Transformer (MMDiT) with rectified-flow training

Stable Diffusion 3.5 Large is the flagship model of Stability AI's October 2024 SD 3.5 release. The architecture is a Multimodal Diffusion Transformer (MMDiT), introduced in the SD3 paper 'Scaling Rectified Flow Transformers for High-Resolution Image Synthesis' (Esser et al. 2024). MMDiT processes text and image tokens with separate parameter streams that interact only at the attention layers, which improves text fidelity and prompt adherence compared with classical cross-attention. The 8.1B parameter backbone is trained with the rectified-flow / flow-matching objective on a curated image-text dataset, with classifier-free guidance distilled into the model. SD 3.5 Large uses three parallel text encoders — CLIP-L, CLIP-G and T5-XXL — so prompts up to ~256 tokens are richly understood. Inference runs at 1024x1024 or higher with 28-50 sampling steps. SD 3.5 Large improves over the original SD3 Medium with better aesthetics, stronger prompt adherence and substantially better photorealism, addressing the widely criticised 'anatomy issues' of SD3 Medium 2B.

Parameters: 8.1B parameters
Context: 256 tokens

What it can do

Open weights with permissive Community License (commercial up to $1M revenue)
8.1B parameter MMDiT — largest open Stability model
Three text encoders (CLIP-L, CLIP-G, T5-XXL) for strong prompt adherence
1024x1024 native + scalable to higher resolutions
Strong photorealism and aesthetic control
Compatible with ControlNet, LoRA, IP-Adapter ecosystem
Self-hostable on a single 24GB+ GPU with quantisation
Best for: open creative pipelines, fine-tuning, ControlNet workflows, research, on-prem deployments.

Training & License

Pretrained on a large licensed and publicly available image-text dataset; data was filtered for safety, deduplication and quality.

License: Stability AI Community License — free for non-commercial use and commercial use up to $1M annual revenue; enterprise license required above that threshold.

Known limitations