Cinematic Masterpieces with Chinook: The Ultimate SDXL Model for Film-Like Images

prompt:

ultra closeup filmgrab portrait of fat monster crazy clown. The face should be sadistic, creepy, eerie, , unshaved, (partial peeling make-up:1.3), dangerous and unstable, intricate textures, dirty clothes, laced embroideries, intricate layered messy makeup, shadow eyes, not a friend, intricate dark circus toys factory background with a lot of creepy dolls and toys, unsettling, film grain, analog, cinematic, rim lighting, realistic, lowkey lighting, (dutch angle style:1.2), high budget, movie still

illustration, sketch, 3d render, glowing eyes, surexposed, saturated
Negative prompt

Introduction to Chinook

In the ever-evolving world of Stable Diffusion models, Chinook stands out as a true cinephile's dream. Created by the talented Ausurety, known for the popular Cheyenne model, Chinook is designed to produce stunningly realistic, cinematic images that look like they've been plucked straight from the silver screen.

Prompt:

cinematic, film grain, RAW, movie still, realistic photography, manifestation black and white fighting, chiaroscuro lighting
Negative prompt: illustration, cartoon, anime, 3d render, painting, crayon, sketch, graphite, impressionist, unreal engine
Steps: 37, Sampler: DPM++ 3M SDE Karras, CFG scale: 4, Seed: 1953958756, Size: 1536x640, Model hash: 59681d5417, Model: CHINOOK_v10, Clip skip: 2, RNG: NV

starting image

ending image

Key Features:

• Photorealistic and fantastical imagery
• Optimized for wide, cinematic formats
• Unique ability to capture mood and narrative in still images

Installing Chinook: Bring Hollywood to Your Desktop

Getting started with Chinook is straightforward:

1. Download the .safetensor file from Civitai
2. Place the file in your /checkpoints folder of your Stable Diffusion interface
3. For easy installation, use Stability Matrix or Diffus en

VAE or No VAE?

Chinook offers two versions:

• VAE Backed: Easier for beginners
• Without VAE: For advanced users who want to experiment with different VAEs

Using Chinook

Recommended Parameters:

• Guidance Scale (CFG): 3-5
• Sampler: DPM++ 2M Karras
• Steps: 30-50
• Optimal Resolution: 1536x640 (panoramic)

Prompt Structure:

Chinook works well with both keyword-based and descriptive prompts. Here's a starter prompt:

cinematic, film grain, RAW, movie still, realistic photography, [SUJET], [ECLAIRAGE], dutch angle style, chromatic aberration
positive

illustration, cartoon, anime, 3d render, painting, crayon, sketch, graphite, impressionist, unreal engine
negative