Expired50% confidenceReleaseExact time
Models that have been converted include Krea2 Turbo, Krea2 Raw, Qwen-Image 2512, Qwen-Image-Edit 2511, and FireRed-Image-Edit 1.1.
1
Sources
50%
Confidence
Medium-term (~90 days)
Relevance
7/12/2026
First Seen
Valid until: 10/10/2026(expired)
Sources
INT4 ConvRot W4A4 Quantization: Running Large Image Models on Low-VRAM GPUs in ComfyUI
redditr/comfyui7/11/2026
Related Claims
UnverifiedSome developers have converted Krea2 Turbo, Krea2 Raw, Qwen-Image 2512, Qwen-Image-Edit 2511, and FireRed-Image-Edit 1.1 to INT4 ConvRot W4A4 format and adapted them for the latest version of ComfyUI.85% similarUnverifiedThis test covered six models: Krea, Ideogram, Flux 1 Dev, Flux 2 Dev, Flux 2 Klein, and Z Image.67% similarUnverifiedKrea 2 Turbo is an image model that prioritizes generation speed.62% similarUnverifiedThe Qwen3-32B model contains 64 Transformer decoder layers.61% similarUnverifiedQwen Code recently released continuous updates from V12 to V14, introducing feature upgrades such as planning mode, visual intelligence, and Zed editor integration.59% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/490298API
curl https://kongchang.com/api/v1/knowledge/claims/490298MCP
get_claim(id=490298)