Updated 2026-09-20 · ranked by real download data

Commercial-use image models

Ranked nightly from our download snapshots of the Hugging Face catalog. Every entry shows its licence and what hardware it realistically needs.

#ModelParamsContextCommercial use30dMin VRAM
01 Z-Image-Turbo +4 variants Tongyi-MAI · text-to-image 6.2 B ✓ apache-2.0 660 K from 37.5 GB
02 FLUX.1-schnell +2 variants black-forest-labs · text-to-image 11.9 B ✓ apache-2.0 603 K from 65.9 GB
03 Qwen-Image-Lightning lightx2v · text-to-image ✓ apache-2.0 423 K from 72.6 GB
04 Z-Image-Lora +5 variants nphSi · text-to-image ✓ apache-2.0 354 K from 109.6 GB
05 Qwen-Image +2 variants Qwen · text-to-image 20.4 B ✓ apache-2.0 311 K from 67.0 GB
06 Qwen-Image-2512-GGUF +1 variant unsloth · text-to-image ✓ apache-2.0 217 K from 8.6 GB
07 Qwen-Image-2512-Lightning lightx2v · text-to-image ✓ apache-2.0 169 K from 118.7 GB
08 NetaYume-Lumina-Image-2.0 duongve · text-to-image ✓ apache-2.0 140 K from 99.5 GB
09 LCM_Dreamshaper_v7 SimianLuo · text-to-image 860 M ✓ mit 137 K from 10.4 GB
10 controlnet-union-sdxl-1.0 xinsir · text-to-image 1.3 B ✓ apache-2.0 118 K from 6.2 GB
11 controlnet-openpose-sdxl-1.0 xinsir · text-to-image 1.3 B ✓ apache-2.0 109 K from 6.2 GB
12 girlslike-krea2 ifmylove2011 · text-to-image 12.8 B ✓ apache-2.0 82 K from 7.3 GB
13 FLUX.2-klein-4B-mflux-4bit Runpod · text-to-image ✓ apache-2.0 81 K from 5.6 GB
14 nunchaku-qwen-image-edit nunchaku-ai · text-to-image ✓ apache-2.0 77 K from 162.9 GB
15 HiDream-I1-Fast HiDream-ai · text-to-image 17.1 B ✓ mit 70 K from 55.0 GB
16 Chroma1-HD lodestones · text-to-image 8.9 B ✓ apache-2.0 66 K from 51.7 GB
17 Z-Image-Loras SDim1973 · text-to-image ✓ apache-2.0 63 K from 90.1 GB
18 Z-Image +1 variant Tongyi-MAI · text-to-image 6.2 B ✓ apache-2.0 62 K from 24.0 GB
19 InstantID InstantX · text-to-image ✓ apache-2.0 57 K from 3.3 GB
20 girlslike-zimage ifmylove2011 · text-to-image 6.2 B ✓ apache-2.0 55 K from 8.2 GB
21 Flux2-Klein-9B-True-V3 wikeeyang · text-to-image ✓ apache-2.0 53 K from 6.8 GB
22 Chroma-GGUF silveroxides · text-to-image ✓ apache-2.0 50 K from 6.0 GB
23 FLux_Klein_9B_NSFW diroverflo · text-to-image ✓ apache-2.0 48 K from 0.7 GB
24 controlnet-scribble-sdxl-1.0 xinsir · text-to-image 1.3 B ✓ apache-2.0 38 K from 3.4 GB
25 Chroma-GGUF-All-Versions duyntnet · text-to-image ✓ apache-2.0 38 K from 6.0 GB
Membership and ranking refresh nightly after the snapshot run. VRAM is an estimate for the smallest quantization at 8K context — methodology.