PrunaAI dropped Few-step LoRA for Qwen-Image-2.1
- 5/ 8 steps
- 6x faster
- no CFG
- T2I + multi-image editing
huggingface.co/PrunaAI/Pruna…
2
14
139
6,500
Awesome Qwen-Image 2.1
already too much data: models, loras, quants...
Keeping the tradition alive: just curated a list of all the best goodies for Qwen Image
github.com/wildminder/awesom…
Meet Qwen-Image-2.1, the most balanced and cost-effective image generation model in the Qwen-Image series! Now open weights! 🎨
A unified model for both generation and editing, delivering top-tier quality in a lightweight package.
Highlights: 👀
- Compact & exceptionally fast: A lightweight 7B architecture that outperforms most closed-source models, with drastically accelerated inference for multi-image inputs.
- Native transparency: Natively generates and edits RGBA layers, enabling seamless compositing and text editing within transparent images.
- Versatile, high-fidelity editing: Supports up to 10 reference images and precise local control while preserving strict fidelity for portraits and products.
- Broad coverage & stunning aesthetics: Excels at panoramas, infographics, and virtual try-ons, delivering realistic textures and elegant typography.
Start to create your next masterpiece with Qwen-Image-2.1! 🖼️
- Blog: qwen.ai/blog?id=qwen-image-2…
- GitHub: github.com/QwenLM/Qwen-Image…
- Model Scope: modelscope.cn/models/Qwen/Qw…
- Hugging Face: huggingface.co/Qwen/Qwen-Ima…
3
21
286
25,313
Space Bunny knows how to hop.
Works quite well with ZCode: manages workflows and organizes agents properly.
This model will be a real workhorse for daily tasks:
> solid general text processing
> no hallucinations at 300K context
> multimodal
> good programming skills
> not too talkative, spends fewer tokens on thinking
Space Bunny (stealth model) is free for the next week
- 1M Context
- Multi-modal
- Zero Data Retention
3
1,203
Qwen-Image-2.1 Fun Controlnet Union by Alibaba!
massive quality-of-life upgrade! one 7.0 GB checkpoint drives 8 controls:
- Canny/Depth/Grayscale/HED/Lineart/MLSD/Pose/Scribble
- Inpainting
- Super fast
huggingface.co/alibaba-pai/Q…
40
343
18,758
Qwen-Image-2.1 Object Remover Bbox LoRA
Turbo LoRA for bbox-guided object removal.
- strips selected objects while preserving textures, lighting, shadows, perspective
may artifact on complex backgrounds, overlaps, fine details, reflections.
huggingface.co/prithivMLmods…
3
14
140
7,925
Viggle just updated the Qwen‑Image‑2.1 Turbo LoRA to v0.2.
LoRA r256 1.3GB;
diversity 0.93x base, zero composition drift;
beats v0.1 on fidelity/sharpness;
still WIP
huggingface.co/Viggle/Qwen-I…
1
2
96
6,079
LTX-2.5 Multiple Subject Reference V2.
multi-subject video refs via shared latent tokens;
- up to 5 images
- retrieves char/clothing/object/scene details
- preserves multi-characters, objects, backgrounds
+ supports subject-object composition.
huggingface.co/LiconStudio/L…
2
11
93
4,590
Ming Image 0.1 Design is now available in ComfyUI !
huge thanks to Kijai!
huggingface.co/Kijai/Ming-Im…
Impressive gift for UI/UX designers!
Ming-Image Design - 6B open-source T2I for UI, infographics, posters & text-rich designs;
- full compositions with RGBA transparent backgrounds
- pairs with Design-Layer for editable transparent layer decomp;
- 2048 x 2048
+ VLM prompt enhancer.
huggingface.co/inclusionAI/M…
5
28
210
17,993
Impressive gift for UI/UX designers!
Ming-Image Design - 6B open-source T2I for UI, infographics, posters & text-rich designs;
- full compositions with RGBA transparent backgrounds
- pairs with Design-Layer for editable transparent layer decomp;
- 2048 x 2048
+ VLM prompt enhancer.
huggingface.co/inclusionAI/M…
We’re open-sourcing the Ming-Image-0.1-Design family:
• Ming-Image-0.1-Design, 6B
• Ming-Image-0.1-Design-Layer, 6B
• Two open-source Agent Skills: the Ling UI Design Skill and the Image-to-Editable-PPT Skill
Ming-Image-0.1-Design ranks #1 among open-weight models on Artificial Analysis’s UI/UX Design leaderboard. 🧵
1
14
120
25,618
Qwen-Image-2.1 Viggle 4step Turbo LoRA
- for T2I + 1-3-ref instruction editing;
- full 14.2 GB bf16 transformer or 340 MB r64 LoRA
- still lags base on complex multi-ref/face/identity edits
huggingface.co/Viggle/Qwen-I…
2
5
77
5,455
Wild! The Qwen team is currently OpenAI’s biggest open-source headache!
If Qwen 4.0 delivers, the open-weight AI scene gets crazy strong.
> Qwen 4.0 27B - the sweet spot for open-source / running locally on consumer GPUs.
> Flash / Plus / Max - Fast & cheap models up to full frontier-tier beasts.
And the 5–10 Trillion Param Rumor (Qwen 4.5 / 5): that’s insane scale. It means they’re aiming directly for AGI-level compute clusters
Qwen4 全家桶即将到来！
Qwen4-Max
Qwen4-Flash
Qwen4-Plus
还有炙手可热的 Qwen4-27B！！！
现场说未来Qwen 的模型规格会达到 5-10T！
Qwen 加油！干死达里奥！！
3
21
2,058
Tencent dropped Hy Image3.5 preview!
Great improvement and nice quality. Much better than HY Image 3.0 Instruct.
same approach: deep reasoning before editing/creating.
- doesn’t need detailed prompts
- follows instructions correctly
- solid composition understanding
Pushes old image restoration to a new level.
Yeah, results aren’t always perfect, but it doesn’t destroy faces or human features.
Hy Image3.5 preview is live. 🚀
Professional-grade image generation, +30% win rate in human eval vs Hy Image3.0
Both Text to image & Image to image available.
Up to 2K. Better Consistency.
API: console.tencentcloud.com/tok…
Priced for everyone. $0.024 per image on Tencent Cloud API.
We only charge for what we generate — your reference images are free.
Two weeks free (Only in OnSolo and Miora):
Miora miora.design/
OnSolo onsolo.ai/
Try it and tell us where it breaks.
3
9
92
8,042
Minimax H3 1950s Sci-Fi LoRA
That retro-futurism vibe (Forbidden Planet, cheesy classic space stuff) is super fun and stylish.
huggingface.co/neph1/1950sSc…
2
10
90
3,583
ComfyUI-Photoshoot
just made character consistency and photo shoots insanely easy. No more writing 40 prompts by hand.
• Person builder
• contact sheet UI
• instant setups for Editorial, Lookbooks, Headshots, Shoes.
• dramaturgy engine
• auto-queue batching
If you still churn out photos for AdobeStock, this will make your life easier.
github.com/ralksta/ComfyUI-P…
4
35
458
21,651
The ’80s are back!
make your own authentic horror films with MiniMax‑H3!
huggingface.co/neph1/1980s_h…
6
9
250
21,022
And ComfyUI will support Qwen‑Image‑2.1 right at launch. Awesome!
github.com/Comfy-Org/ComfyUI…
Qwen-Image-2.1 is going open source! 🎨
⌛Join the countdown modelscope.cn/models/Qwen/Qw…
Weights + code. Download it, run it, build with it. 💻
2
37
4,481
ComfyUI heatmaps node.
maps which prompt tags shaped each image region
- SDXL only
- Tag Explorer shows per-pixel shares on hover/click
- pin multiple tags; strength/smooth controls
github.com/alchemine/comfyui…
2
5
42
2,326
MiniMax H3 LoRA for cinematic gunfights & close-quarters action
chains ranged firefights to John Wick-style combat with firearms, strikes, throws, struggles
huggingface.co/JOKER141/Mini…
2
4
38
3,200
MoGe‑3 is now available in ComfyUI.
fine-detail monocular 3D geometry from open-domain images
recovers metric point/depth/normal maps + FOV in one pass
huggingface.co/Comfy-Org/MoG…
4
37
306
15,484
Minimax-H3 Third Person View LoRA
If you ever need cinematic game cutscenes…
makes precise third-person/over-the-shoulder tracking with spring-arm motion
+ seamless FPV transitions, native HUD, reticles, QTE prompts, boss bars, action VFX.
huggingface.co/WarmBloodAban…
1
2
68
3,620