We are excited to open-source Qwen-Image-2.1, a unified text-to-image generation and image editing model in the Qwen family. With just 7B parameters in its visual generation component (32 Single-Stream DiT layers), Qwen-Image-2.1 balances generation quality, inference efficiency, and versatility. Four key improvements define this release: - Compact and Efficient — A lightweight architecture with mixed-granularity attention and prefix KV cache reuse delivers strong image quality at low computational cost. - Native Transparency, Unified Creation and Editing — Generate regular or transparent (RGBA) images from text, edit transparent layers, and extract subjects from photographs—all in one…
Open-weight model · Text to image
Qwen-Image-2.1-viggle-turbo
by Viggle AI Viggle/Qwen-Image-2.1-viggle-turbo
Qwen-Image-2.1-viggle-turbo is an open-weight model for text to image from Viggle AI, released under other. It has 7.1B parameters. At 16-bit it needs about 17.1 GB of GPU memory, which fits on 1x MI300X from $1.85 an hour, at the lowest prices in the SAVRN Index. It draws 12.4k downloads a month.
Built with Qwen. A few-step distilled student of Qwen/Qwen-Image-2.1, trained by Viggle with Distribution Matching Distillation.
Parameters7.1B
Context—
Weights24.8 GB
Licenseother
AccessOpen weights
Monthly Downloads12.4k
Runs On
What it takes to serve Qwen-Image-2.1-viggle-turbo (7.1B parameters): the memory its weights need at each precision, and the cheapest way to rent enough data-center GPUs to hold them.
| Precision | Weights | Memory needed | Cheapest setup | Per hour | Also fits |
|---|---|---|---|---|---|
| 16-bit | 14.2 GB | 17.1 GB | 1x MI300X (192 GB) Vultr |
$1.85 | 1x H100 $1.99 · 1x MI325X $2.00 |
| 8-bit | 7.1 GB | 8.5 GB | 1x MI300X (192 GB) Vultr |
$1.85 | 1x H100 $1.99 · 1x MI325X $2.00 |
| 4-bit | 3.6 GB | 4.3 GB | 1x MI300X (192 GB) Vultr |
$1.85 | 1x H100 $1.99 · 1x MI325X $2.00 |
Memory is the weights at that precision plus 20% for the runtime and a short context; a long context needs more. Prices are the lowest on-demand hourly rates in the SAVRN Index, read Oct 16, 2026.
Qwen-Image-2.1-viggle-turbo on every accelerator the SAVRN Index prices, at every precision
Model Card
Built with Qwen. A few-step distilled student of Qwen/Qwen-Image-2.1, trained by Viggle with Distribution Matching Distillation. It does both text-to-image and instruction-driven editing with 1–3 reference images in 6 transformer passes instead of 40, with no classifier-free guidance. src="https://huggingface.co/Viggle/Qwen-Image-2.1-viggle-turbo/resolve/main/assets/viggleturbopromo.mp4"> About 5× faster than the 40-step base model end to end, and very competitive with it in quality: on the official Qwen examples the two are hard to tell apart on most prompts, and on some — dense layouts like the character style infographic — we prefer the turbo's render. See for yourself in the Comparison…
Excerpt from the card by Viggle AI, licensed other.
Identity and Version
- Repository
- Viggle/Qwen-Image-2.1-viggle-turbo
- Publisher
- Viggle AI
- Task
- Text to image
- Modality
- Image
- Library
- diffusers
- Parameters
- 7.1B parameters
- Languages
- dmd, few-step
- Revision
- d8c745bcb339c0801e21eecdc1947c5ba9e18b2e
- First published
- 2026-09-22
- Last updated
- 2026-09-25
Files and Weights
24 files, 24.8 GB in total. The weights are 9 files totalling 24.8 GB in safetensors.
Weights9 files · 24.8 GB
Configuration8 files · 179.6 KB
Documentation3 files · 30.6 KB
Other3 files · 14.8 MB
Repository1 file · 1.6 KB
Every file
| File | Type | Size | SHA-256 |
|---|---|---|---|
| Qwen-Image-2.1-viggle-turbo-4step-lora-r64.safetensors | Weights | 339.8 MB | 5c494dce662a |
| Qwen-Image-2.1-viggle-turbo-v0.2-5step-lora-r128.safetensors | Weights | 679.6 MB | 7096a791d0f1 |
| Qwen-Image-2.1-viggle-turbo-v0.2-5step-lora-r256.safetensors | Weights | 1.4 GB | 7fd196e14629 |
| Qwen-Image-2.1-viggle-turbo-v0.2.1-6step-lora-r128.safetensors | Weights | 679.6 MB | bafb91d0047d |
| Qwen-Image-2.1-viggle-turbo-v0.2.1-6step-lora-r256.safetensors | Weights | 1.4 GB | 2a0148f5c73a |
| peft/adapter_model.safetensors | Weights | 679.6 MB | c7d944042f50 |
| peft_v0.2.1/adapter_model.safetensors | Weights | 2.7 GB | 50e6436ecc4b |
| peft_v0.2/adapter_model.safetensors | Weights | 2.7 GB | 9845b0da828a |
| transformer/diffusion_pytorch_model.safetensors | Weights | 14.2 GB | a011a284b2dc |
| comfyui/Qwen-Image-2.1-viggle-turbo-edit.json | Configuration | 98.2 KB | — |
| comfyui/Qwen-Image-2.1-viggle-turbo-t2i.json | Configuration | 71.8 KB | — |
| comfyui/viggle_turbo.py | Configuration | 5.5 KB | — |
| peft/adapter_config.json | Configuration | 1.1 KB | — |
| peft_v0.2.1/adapter_config.json | Configuration | 1.1 KB | — |
| peft_v0.2/adapter_config.json | Configuration | 1.1 KB | — |
| scheduler/scheduler_config.json | Configuration | 484 B | — |
| transformer/config.json | Configuration | 369 B | — |
| LICENSE | Documentation | 7.8 KB | — |
| NOTICE | Documentation | 2.1 KB | — |
| README.md | Documentation | 20.6 KB | — |
| assets/viggle_turbo_promo.mp4 | Other | 14.7 MB | fdb14e6b1f32 |
| comfyui/input/cat.webp | Other | 46.2 KB | — |
| comfyui/input/woman2.webp | Other | 83.7 KB | — |
| .gitattributes | Repository | 1.6 KB | — |
License and Download
- License
- other
- Access
- Open weights, no gate
- Download size
- 24.8 GB
Download from Viggle AI
Released by Viggle AI through its official repository on Hugging Face.
Built From
- Adapter of Qwen/Qwen-Image-2.1
- Derived from Qwen/Qwen-Image-2.1
Memory Requirements
| Precision | Weights in memory |
|---|---|
| As published | 24.8 GB |
| 16-bit | 14.2 GB |
| 8-bit | 7.1 GB |
| 4-bit | 3.6 GB |
Weights only, from the published parameter count; the key-value cache and runtime add to this.
Questions About Qwen-Image-2.1-viggle-turbo
How much GPU memory does Qwen-Image-2.1-viggle-turbo need?
About 17.1 GB at 16-bit and 4.3 GB at 4-bit: the weights (7.1B parameters) plus a working margin. A long context needs more.
What is the cheapest GPU to run Qwen-Image-2.1-viggle-turbo on?
At 16-bit, 1x MI300X from $1.85 an hour; at 4-bit, 1x MI300X from $1.85 an hour, at the lowest on-demand prices the SAVRN Index lists.
What license is Qwen-Image-2.1-viggle-turbo released under?
other, as its publisher declares it. Read the license text before commercial use.
Similar Models
Qwen-Image-2.1-Turbo is an accelerated checkpoint of Qwen-Image-2.1 for text-to-image generation and image editing with 8 denoising steps. It uses the same 7B visual generation architecture and loads directly with QwenImage21Pipeline in Diffusers. The checkpoint includes its recommended sampling schedule, so it is ready to use without manually configuring the scheduler. Generation uses CFG=1 by default, and prefix KV caching reuses the text and reference-image context across denoising steps. For more details, see the GitHub repo and Qwen-Image-2.1 blog. Install a CUDA-compatible PyTorch build, then install the latest Diffusers source and the following dependencies: This checkpoint requires…
Welcome to the official repository for the Z-Image（造相）project! Z-Image is a powerful and highly efficient image generation model family with 6B parameters. Currently there are four variants: - Z-Image-Turbo – A distilled version of Z-Image that matches or exceeds leading competitors with only 8 NFEs (Number of Function Evaluations). It offers sub-second inference latency on enterprise-grade H800 GPUs and fits comfortably within 16G VRAM consumer devices. It excels in photorealistic image generation, bilingual text rendering (English & Chinese), and robust instruction adherence. - Z-Image – The foundation model behind Z-Image-Turbo. Z-Image focuses on high-quality generation, rich…
Stable Diffusion 3.5 Large is a Multimodal Diffusion Transformer (MMDiT) text-to-image model that features improved performance in image quality, typography, complex prompt understanding, and resource-efficiency. Please note: This model is released under the Stability Community License. Visit Stability AI to learn or contact us for commercial licensing details. - For individuals and organizations with annual revenue above $1M: please contact us to get an Enterprise License. For local or self-hosted use, we recommend ComfyUI for node-based UI inference, or diffusers or GitHub for programmatic use. - Text Encoders： This model was trained on a wide variety of data, including synthetic data and…
The text-to-image prompt rewriter for Qwen-Image-2.1, a fine-tuned Qwen3.5-VL 9B that turns a short request into a detailed English prompt plus a recommended aspect ratio, with its refusal behaviour reduced by Heretic directional ablation. bf16, same shapes and parameter count as the source. systemprompt.txt is included and required (the unmodified file from the source repo). for image edits. Measured by Heretic on mlabonne/harmfulbehaviors (refusals) and mlabonne/harmlessalpaca (KL divergence, the damage to ordinary behaviour), with Heretic's default system prompt. The first row is an independent evaluation of the exported weights (evaluatemodel); the search's own figure for the same trial…
NVFP4 (4-bit floating point, W4A4) build of darrellbest/Qwen-Image-2.1-PE-T2I-Heretic, the refusal-ablated text-to-image prompt rewriter for Qwen-Image-2.1. For vLLM on NVIDIA Blackwell GPUs, which run NVFP4 natively. 11 GB instead of 18 GB. systemprompt.txt is included and required, exactly as for the original. The linear-attention layers carry a recurrent state and the vision tower encodes the input image; both were left in bf16, as other quantizations of this model family do. That is why the file is 11 GB rather than ~6 GB. Made with llm-compressor 0.13.0 (QuantizationModifier, scheme="NVFP4"), calibrated on 64 samples in the model's real input format: its own system prompt and a short…