Parameter GuideStable Diffusion

Stable Diffusion Sampler Guide

The sampler setting in Stable Diffusion controls the algorithm used to convert noise into a finished image. This guide compares the most common samplers — including Euler, DPM++, and DDIM — so you can pick the right one for your workflow and generation goals.

View in other languages

Bahasa IndonesiaPengaturan sampler di Stable Diffusion mengontrol algoritma yang digunakan untuk mengubah noise menjadi gambar yang sudah jadi. Panduan ini membandingkan sampler paling umum — termasuk Euler, DPM++, dan DDIM — sehingga Anda dapat memilih yang tepat untuk alur kerja dan tujuan generasi Anda.

日本語Stable Diffusionのサンプラー設定は、ノイズを完成した画像に変換するために使用されるアルゴリズムを制御します。このガイドでは、Euler、DPM++、DDIMなどの最も一般的なサンプラーを比較し、ワークフローと生成目標に合った適切なサンプラーを選べるようにします。

한국어Stable Diffusion의 샘플러 설정은 노이즈를 완성된 이미지로 변환하는 데 사용되는 알고리즘을 제어합니다. 이 가이드는 Euler, DPM++, DDIM 등 가장 일반적인 샘플러를 비교하여 워크플로우와 생성 목표에 맞는 올바른 샘플러를 선택할 수 있도록 도와줍니다.

PolskiUstawienie samplera w Stable Diffusion kontroluje algorytm używany do przekształcania szumu w gotowy obraz. Ten przewodnik porównuje najpopularniejsze samplery — w tym Euler, DPM++ i DDIM — abyś mógł wybrać odpowiedni do swojego przepływu pracy i celów generowania.

PortuguêsA configuração de sampler no Stable Diffusion controla o algoritmo usado para converter ruído em uma imagem finalizada. Este guia compara os samplers mais comuns — incluindo Euler, DPM++ e DDIM — para que você possa escolher o mais adequado para seu fluxo de trabalho e objetivos de geração.

РусскийНастройка сэмплера в Stable Diffusion управляет алгоритмом, который преобразует шум в готовое изображение. В этом руководстве сравниваются наиболее распространённые сэмплеры — Euler, DPM++ и DDIM — чтобы вы могли выбрать подходящий для своего рабочего процесса.

TürkçeStable Diffusion'daki örnekleyici ayarı, gürültüyü tamamlanmış bir görüntüye dönüştürmek için kullanılan algoritmayı kontrol eder. Bu rehber, en yaygın örnekleyicileri — Euler, DPM++ ve DDIM dahil — karşılaştırarak iş akışınız ve üretim hedefleriniz için doğru olanı seçmenizi sağlar.

中文Stable Diffusion 中的采样器设置控制将噪声转换为成品图像所使用的算法。本指南比较了最常见的采样器——包括 Euler、DPM++ 和 DDIM——帮助你为工作流程和生成目标选择合适的采样器。

Values & Usage

Sampler: LowLow

More variation and creative interpretation in the output

Sampler: Medium (default)Medium (default)

Balanced results suitable for most use cases

Sampler: HighHigh

More consistent and detailed output, closer to prompt intent

Sampler: MaximumMaximum

Highest fidelity to prompt — best for professional and final output

Example Prompts

Demonstrating Sampler at Low — More variation and creative interpretation in the output

beginner
(ultra-detailed:1.3), vaporwave scenic composition, landscape, candlelight warmth, cool blue-silver palette, sharp focus, intricate details, professional

Negative prompt: lowres, bad anatomy, bad hands, text, error, worst quality, low quality, blurry, deformed, disfigured, watermark, signature

Demonstrating Sampler at Medium (default) — Balanced results suitable for most use cases

beginner
(highly detailed:1.3), minimalist scenic composition, landscape, dramatic side lighting, neon accent colors, sharp focus, intricate details, professional

Negative prompt: lowres, bad anatomy, bad hands, text, error, worst quality, low quality, blurry, deformed, disfigured, watermark, signature

Demonstrating Sampler at High — More consistent and detailed output, closer to prompt intent

intermediate
(best quality:1.3), gothic scenic composition, landscape, harsh directional light, muted pastel shades, sharp focus, intricate details, professional

Negative prompt: lowres, bad anatomy, bad hands, text, error, worst quality, low quality, blurry, deformed, disfigured, watermark, signature

Demonstrating Sampler at Maximum — Highest fidelity to prompt — best for professional and final output

advanced
(masterpiece:1.3), ukiyo-e scenic composition, landscape, sunset rim light, warm earth tones, sharp focus, intricate details, professional

Negative prompt: lowres, bad anatomy, bad hands, text, error, worst quality, low quality, blurry, deformed, disfigured, watermark, signature

Pro Tips

  • 1Start with the default Sampler value and make small adjustments to understand how it affects your specific prompt style.
  • 2Different styles respond differently to Sampler changes — what works for photorealistic may not work for anime or abstract.
  • 3Keep notes on which Sampler values work best for your favorite prompt structures — this builds your personal Stable Diffusion knowledge base.
  • 4When sharing prompts, always include your Sampler setting — it can dramatically change results even with identical prompt text.
  • 5For batch generation, lock your Sampler value to maintain consistency across an entire set of related images.

Common Mistakes

Using extreme Sampler values without understanding the trade-offs — moderate values usually produce the most balanced results.
Assuming the same Sampler value works for all styles — always test with your specific prompt before committing to a batch.
Changing Sampler and other settings simultaneously — adjust one parameter at a time to understand each one's effect.

Frequently Asked Questions

Free to try — no credit card required

Optimize Your Prompts with AI

Stop guessing. Geniea analyzes and optimizes your Stable Diffusion prompts in seconds for better results every time.

Try Geniea Free