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…
Open-weight model · Text to image
Qwen-Image-2.1-PE-T2I-Heretic
by Darrellbest darrellbest/Qwen-Image-2.1-PE-T2I-Heretic
Qwen-Image-2.1-PE-T2I-Heretic is an open-weight model for text to image from Darrellbest, released under other. It has 9.4B parameters and a 262,144-token context. At 16-bit it needs about 22.6 GB of GPU memory, which fits on 1x MI300X from $1.85 an hour, at the lowest prices in the SAVRN Index. It draws 39 downloads a month.
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.
Parameters9.4B
Context262,144
Weights18.8 GB
Licenseother
AccessOpen weights
Monthly Downloads39
Runs On
What it takes to serve Qwen-Image-2.1-PE-T2I-Heretic (9.4B 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 | 18.8 GB | 22.6 GB | 1x MI300X (192 GB) Vultr |
$1.85 | 1x H100 $1.99 · 1x MI325X $2.00 |
| 8-bit | 9.4 GB | 11.3 GB | 1x MI300X (192 GB) Vultr |
$1.85 | 1x H100 $1.99 · 1x MI325X $2.00 |
| 4-bit | 4.7 GB | 5.6 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 Sep 24, 2026.
Qwen-Image-2.1-PE-T2I-Heretic on every accelerator the SAVRN Index prices, at every precision
Model Card
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…
Excerpt from the card by Darrellbest, licensed other.
Configuration
- Architecture
- Qwen3_5ForConditionalGeneration
- Context length (tokens)
- 262,144
- Layers
- 32
- Hidden size
- 4,096
- Feed-forward size
- 12,288
- Attention heads
- 16
- Key/value heads
- 4
- Head dimension
- 256
- Vocabulary size
- 248,320
- Model type
- qwen3_5
Identity and Version
- Repository
- darrellbest/Qwen-Image-2.1-PE-T2I-Heretic
- Publisher
- Darrellbest
- Task
- Text to image
- Modality
- Image
- Library
- Not stated by the source
- Parameters
- 9.4B parameters
- Languages
- Not stated by the source
- Revision
- d6a96e5398fae3c64c97b528606da2bd6252b2e9
- First published
- 2026-09-22
- Last updated
- 2026-09-24
Files and Weights
15 files, 18.8 GB in total. The weights are 4 files totalling 18.8 GB in safetensors.
Weights4 files · 18.8 GB
Configuration4 files · 73.7 KB
Tokenizer2 files · 20.0 MB
Documentation2 files · 13.2 KB
Other2 files · 17.8 KB
Repository1 file · 1.6 KB
Every file
| File | Type | Size | SHA-256 |
|---|---|---|---|
| model-00001-of-00004.safetensors | Weights | 4.9 GB | 88257291b229 |
| model-00002-of-00004.safetensors | Weights | 5.0 GB | bccde99819ce |
| model-00003-of-00004.safetensors | Weights | 5.0 GB | 4f84fa4e457e |
| model-00004-of-00004.safetensors | Weights | 3.9 GB | 526181eec8bd |
| config.json | Configuration | 3.0 KB | — |
| generation_config.json | Configuration | 205 B | — |
| model.safetensors.index.json | Configuration | 69.3 KB | — |
| processor_config.json | Configuration | 1.3 KB | — |
| LICENSE | Documentation | 7.8 KB | — |
| README.md | Documentation | 5.3 KB | — |
| chat_template.jinja | Other | 7.8 KB | — |
| system_prompt.txt | Other | 10.0 KB | — |
| .gitattributes | Repository | 1.6 KB | — |
| tokenizer.json | Tokenizer | 20.0 MB | 87a7830d63fc |
| tokenizer_config.json | Tokenizer | 1.2 KB | — |
License and Download
- License
- other
- Access
- Open weights, no gate
- Download size
- 18.8 GB
Download from Darrellbest
Released by Darrellbest through its official repository on Hugging Face.
Built From
- Derived from Qwen/Qwen-Image-2.1-PE-T2I
Memory Requirements
| Precision | Weights in memory |
|---|---|
| As published | 18.8 GB |
| 16-bit | 18.8 GB |
| 8-bit | 9.4 GB |
| 4-bit | 4.7 GB |
Weights only, from the published parameter count; the key-value cache and runtime add to this.
Built on This Model
- Quantized fromQwen-Image-2.1-PE-T2I-Heretic-NVFP4
- Derived fromQwen-Image-2.1-PE-T2I-Heretic-NVFP4
- Quantized fromQwen-Image-2.1-PE-T2I-Heretic-GGUF
- Derived fromQwen-Image-2.1-PE-T2I-Heretic-GGUF
Questions About Qwen-Image-2.1-PE-T2I-Heretic
How much GPU memory does Qwen-Image-2.1-PE-T2I-Heretic need?
About 22.6 GB at 16-bit and 5.6 GB at 4-bit: the weights (9.4B parameters) plus a working margin. A long context needs more.
What is the cheapest GPU to run Qwen-Image-2.1-PE-T2I-Heretic 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-PE-T2I-Heretic released under?
other, as its publisher declares it. Read the license text before commercial use.
What is Qwen-Image-2.1-PE-T2I-Heretic's context length?
262,144 tokens, from the maximum position embeddings in its published configuration.
Similar Models
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…
FLUX.1 [schnell] is a 12 billion parameter rectified flow transformer capable of generating images from text descriptions. For more information, please read our blog post. 1. Cutting-edge output quality and competitive prompt following, matching the performance of closed source alternatives. 2. Trained using latent adversarial diffusion distillation, FLUX.1 [schnell] can generate high-quality images in only 1 to 4 steps. 3. Released under the apache-2.0 licence, the model can be used for personal, scientific, and commercial purposes. We provide a reference implementation of FLUX.1 [schnell], as well as sampling code, in a dedicated github repository. Developers and creatives looking to…
FLUX.1 [dev] is a 12 billion parameter rectified flow transformer capable of generating images from text descriptions. For more information, please read our blog post. 1. Cutting-edge output quality, second only to our state-of-the-art model FLUX.1 [pro]. 2. Competitive prompt following, matching the performance of closed source alternatives. 3. Trained using guidance distillation, making FLUX.1 [dev] more efficient. 4. Open weights to drive new scientific research, and empower artists to develop innovative workflows. 5. Generated outputs can be used for personal, scientific, and commercial purposes as described in the [FLUX.1 [dev] Non-Commercial…
NSFW Weights for this model are available in Safetensors format. Download them in the Files & versions tab.
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…