Patreon
Home / Apps / Research, Comparisons and Experiments / BF16 vs GGUF Q8 vs FP8 Scaled vs NVFP4: Quality and Speed Co

BF16 vs GGUF Q8 vs FP8 Scaled vs NVFP4: Quality and Speed Comparison

BF16, GGUF Q8, FP8 Scaled and NVFP4 precisions compared for speed and quality in ComfyUI with CUDA 13; NVFP4 ran 87% faster than GGUF Q8 on Z Image Turbo.

comparisonimage generationcomfyuitutorial
BF16 vs GGUF Q8 vs FP8 Scaled vs NVFP4: Quality and Speed Comparison

Highlights

Tested with CUDA 13 and Sage Attention on Z Image Turbo, FLUX Dev, FLUX SRPO, FLUX Kontext and FLUX 2, with the difference in quality shown side by side. Also introduces our NVFP4 model quantizer and FP8 Scaled quantization apps.

Video tutorial

Stop Using GGUF for FLUX Dev: FP8 Scaled and NVFP4 Run Up to 2x Faster (ComfyUI CUDA 13 Test) · chapters and summary

More videos: Qwen Image Models Training - 0 to Hero Level Tutorial - LoRA & Fine Tuning - Base & Edit Model, How to Use SwarmUI Presets & Workflows in ComfyUI + Custom Model Paths Setup for ComfyUI & SwarmUI, AI Images Look Fake? This Realism LoRA Makes Qwen Image Look Like Real Photos

Where to get it: the full post is free on this Patreon post by Furkan Gözükara (SECourses). Published Jan 17, 2026.

Related

Qwen Image 2512 Training Comparison: The Best Model to Train

Qwen Image 2512 Training Comparison: The Best Model to Train

Qwen Image 2512 is the new king for training: the video compares trained Base, 2512, Edit 2509 and Edit 2511 models and shows a big leap in realism.

NVFP4 With CUDA 13 Tutorial: 100%+ Speed Gain & Quality Comparison

NVFP4 With CUDA 13 Tutorial: 100%+ Speed Gain & Quality Comparison

NVFP4 models run over 100% faster in ComfyUI and SwarmUI with CUDA 13; this full tutorial compares NVFP4 quality on FLUX 2, Z Image Turbo and FLUX 1 in grids.

Qwen Image vs FLUX Dev vs FLUX Krea: New Text-to-Image King Comparison

Qwen Image vs FLUX Dev vs FLUX Krea: New Text-to-Image King Comparison

Qwen Image compared with FLUX Dev and FLUX Krea over 700 generations; Qwen follows prompts far better and has a 68% bigger native resolution than FLUX Dev.

More in Research, Comparisons and Experiments.