Skip to content
Agientry
← Back to Videos

Viggle turbo: Qwen-Image 2.1 in 6 steps, 5x faster

Loading video
Loading video

Viggle's turbo student distills Qwen-Image 2.1 from 40 steps to 6 transformer passes, a 5x speedup with no classifier-free guidance, keeping 1-3 reference-image instruction editing; ComfyUI's unmerged node load avoids bf16 truncation loss.

Image GenerationDiffusion modelsDistillationComfyUIQwen
Category: world-model
Author: @Lonely__MH
Date: 2026-09-29T00:00:00
Duration: 33.766s
Source: https://x.com/Lonely__MH/status/2104724839428981098
Reference: https://huggingface.co/Viggle/Qwen-Image-2.1-viggle-turbo

Related Videos

Qwen-Image 2.1 Move: Move and Resize Objects Across Images
0:13

Qwen-Image 2.1 Move: Move and Resize Objects Across Images

@linoy_tsaban · 149 views · 2026-09-29
Image EditingImage GenerationQwen
OmniChar: One Character Across Every Model, Carried in a Portable .char File
0:23

OmniChar: One Character Across Every Model, Carried in a Portable .char File

@aiaicreate · 261 views · 2026-09-28
OmniCharCharacter ConsistencyComfyUI
Whip-pan transitions: 11 shots in one 30-second AI video
0:35

Whip-pan transitions: 11 shots in one 30-second AI video

@Promptwhat · 116 views · 2026-09-28
Video GenerationPromptingWhip Pan
CWM-1: The Most Controllable Interactive Avatar Model
0:45

CWM-1: The Most Controllable Interactive Avatar Model

@LemonSliceAI · 211 views · 2026-09-23
World ModelsVideo GenerationDiffusion models
XPENG XPACE: A Unified Embodied World Model for IRON
1:31

XPENG XPACE: A Unified Embodied World Model for IRON

@CyberRobooo · 263 views · 2026-09-17
World ModelsEmbodied AIXpeng
XGEN Generative World Simulation: JING Egocentric Experience Model Plus DAO Shared-World Engine
1:06

XGEN Generative World Simulation: JING Egocentric Experience Model Plus DAO Shared-World Engine

@XGEN_labs · 526 views · 2026-09-17
World ModelsVideo GenerationInteractive Video