# stable-diffusion-v1-4 by CompVis: Open-Weight Model Source: https://savrn.com/models/stable-diffusion-v1-4 Markdown alternate of the page above; the site index is https://savrn.com/llms.txt --- ## Runs On What it takes to serve stable-diffusion-v1-4 (860M 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 | 1.7 GB | 2.1 GB | 1x [MI300X](https://savrn.com/ai-index/pricing/gpus/mi300x) (192 GB) Vultr | $1.85 | [1x H100](https://savrn.com/ai-index/pricing/gpus/h100) $1.99 · [1x MI325X](https://savrn.com/ai-index/pricing/gpus/mi325x) $2.00 | | 8-bit | 0.9 GB | 1.0 GB | 1x [MI300X](https://savrn.com/ai-index/pricing/gpus/mi300x) (192 GB) Vultr | $1.85 | [1x H100](https://savrn.com/ai-index/pricing/gpus/h100) $1.99 · [1x MI325X](https://savrn.com/ai-index/pricing/gpus/mi325x) $2.00 | | 4-bit | 0.4 GB | 0.5 GB | 1x [MI300X](https://savrn.com/ai-index/pricing/gpus/mi300x) (192 GB) Vultr | $1.85 | [1x H100](https://savrn.com/ai-index/pricing/gpus/h100) $1.99 · [1x MI325X](https://savrn.com/ai-index/pricing/gpus/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](https://savrn.com/ai-index/pricing/gpus), read Oct 25, 2026. [stable-diffusion-v1-4 on every accelerator the SAVRN Index prices, at every precision](https://savrn.com/models/stable-diffusion-v1-4/gpus) ## SAVRN's Notes on stable-diffusion-v1-4 Stable Diffusion 1.4 is the release that brought open image generation to everyone in 2022, and more than 7,000 likes reflect that history. It has 860M parameters and runs on almost any GPU. The CreativeML OpenRAIL-M license allows commercial use with use restrictions that pass on to your outputs and users. For new work, much newer models in our catalog are far better; keep this one for research and reproducibility. SAVRN Research, 2026-09-18 ## Model Card By CompVis, published under creativeml-openrail-m, revision 133a221b8aa7. ### Stable Diffusion v1-4 Model Card Stable Diffusion is a latent text-to-image diffusion model capable of generating photo-realistic images given any text input. For more information about how Stable Diffusion functions, please have a look at ['s Stable Diffusion with Diffusers blog](https://huggingface.co/blog/stable_diffusion). The Stable-Diffusion-v1-4 checkpoint was initialized with the weights of the Stable-Diffusion-v1-2 checkpoint and subsequently fine-tuned on 225k steps at resolution 512x512 on "laion-aesthetics v2 5+" and 10% dropping of the text-conditioning to improve [classifier-free guidance sampling](https://savrn.com/papers/classifier-free-diffusion-guidance). This weights here are intended to be used with the Diffusers library. If you are looking for the weights to be loaded into the CompVis Stable Diffusion codebase,[come here](https://huggingface.co/CompVis/stable-diffusion-v-1-4-original) ### Model Details [Read the full model card (1,835 words)](https://savrn.com/models/stable-diffusion-v1-4/card) ## Identity and Version Repository CompVis/stable-diffusion-v1-4 Publisher CompVis Task Text to image Modality Image Library diffusers Parameters 860M parameters Languages Not stated by the source Revision 133a221b8aa7292a167afc5127cb63fb5005638b First published 2022-08-20 Last updated 2023-08-23 ## Files and Weights 33 files, 23.3 GB in total. The weights are 18 files totalling 23.3 GB in bin, safetensors. Weights18 files · 23.3 GB Configuration9 files · 8.3 KB Tokenizer3 files · 1.6 MB Documentation1 file · 17.1 KB Other1 file · 71.2 KB Repository1 file · 1.4 KB Every file | File | Type | Size | SHA-256 | | --- | --- | --- | --- | | safety_checker/model.fp16.safetensors | Weights | 608.0 MB | 08902f19b1cf | | safety_checker/model.safetensors | Weights | 1.2 GB | 9d6a233ff6fd | | safety_checker/pytorch_model.bin | Weights | 1.2 GB | 193490b58ef6 | | safety_checker/pytorch_model.fp16.bin | Weights | 608.1 MB | 22ba87205445 | | text_encoder/model.fp16.safetensors | Weights | 246.1 MB | 77795e2023ad | | text_encoder/model.safetensors | Weights | 492.3 MB | 7b3a12df205c | | text_encoder/pytorch_model.bin | Weights | 492.3 MB | 770a47a9ffdc | | text_encoder/pytorch_model.fp16.bin | Weights | 246.2 MB | 05eee911f195 | | unet/diffusion_pytorch_model.bin | Weights | 3.4 GB | 62d48b4d841a | | unet/diffusion_pytorch_model.fp16.bin | Weights | 1.7 GB | 3ac986370f51 | | unet/diffusion_pytorch_model.fp16.safetensors | Weights | 1.7 GB | a35404d03ec8 | | unet/diffusion_pytorch_model.non_ema.bin | Weights | 3.4 GB | f5f12f7078f3 | | unet/diffusion_pytorch_model.non_ema.safetensors | Weights | 3.4 GB | ac3d1bff03f9 | | unet/diffusion_pytorch_model.safetensors | Weights | 3.4 GB | 145a07e0f05e | | vae/diffusion_pytorch_model.bin | Weights | 334.7 MB | 1b134cded8eb | | vae/diffusion_pytorch_model.fp16.bin | Weights | 167.4 MB | b7643b3e40b9 | | vae/diffusion_pytorch_model.fp16.safetensors | Weights | 167.3 MB | 4fbcf0ebe55a | | vae/diffusion_pytorch_model.safetensors | Weights | 334.6 MB | a2b5134f4dbc | | feature_extractor/preprocessor_config.json | Configuration | 342 B | — | | model_index.json | Configuration | 541 B | — | | safety_checker/config.json | Configuration | 4.6 KB | — | | scheduler/.ipynb_checkpoints/scheduler_config-checkpoint.json | Configuration | 209 B | — | | scheduler/scheduler_config.json | Configuration | 313 B | — | | text_encoder/config.json | Configuration | 592 B | — | | tokenizer/special_tokens_map.json | Configuration | 472 B | — | | unet/config.json | Configuration | 743 B | — | | vae/config.json | Configuration | 551 B | — | | README.md | Documentation | 17.1 KB | — | | v1-variants-scores.jpg | Other | 71.2 KB | — | | .gitattributes | Repository | 1.4 KB | — | | tokenizer/merges.txt | Tokenizer | 524.6 KB | — | | tokenizer/tokenizer_config.json | Tokenizer | 806 B | — | | tokenizer/vocab.json | Tokenizer | 1.1 MB | — | ## License and Download License creativeml-openrail-m Access Open weights, no gate Download size 23.3 GB [Download from CompVis](https://huggingface.co/CompVis/stable-diffusion-v1-4) Released by CompVis through its official repository on Hugging Face. ## Built From - Described by [arXiv:1910.09700](https://savrn.com/papers/quantifying-the-carbon-emissions-of-machine-learning) - Described by [arXiv:2103.00020](https://savrn.com/papers/learning-transferable-visual-models-from-natural-language-supervision) - Described by [arXiv:2112.10752](https://savrn.com/papers/high-resolution-image-synthesis-with-latent-diffusion-models) - Described by [arXiv:2205.11487](https://savrn.com/papers/photorealistic-text-to-image-diffusion-models-with-deep-language-understanding) - Described by [arXiv:2207.12598](https://savrn.com/papers/classifier-free-diffusion-guidance) ## Memory Requirements | Precision | Weights in memory | | --- | --- | | As published | 23.3 GB | | 16-bit | 1.7 GB | | 8-bit | 0.9 GB | | 4-bit | 0.4 GB | Weights only, from the published parameter count; the key-value cache and runtime add to this. ## Questions About stable-diffusion-v1-4 ### How much GPU memory does stable-diffusion-v1-4 need? About 2.1 GB at 16-bit and 0.5 GB at 4-bit: the weights (860M parameters) plus a working margin. A long context needs more. ### What is the cheapest GPU to run stable-diffusion-v1-4 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. ### Can I use stable-diffusion-v1-4 commercially? Yes, with conditions. stable-diffusion-v1-4 is released under CreativeML Open RAIL-M. CreativeML Open RAIL-M permits use, including commercial use, subject to the use-based restrictions in its attachment, which must be passed on downstream. ## Similar Models Model · Text to image ### [stable-diffusion-v1-5](https://savrn.com/models/stable-diffusion-v1-5) [SD v1.5](https://savrn.com/model-publishers/stable-diffusion-v1-5) Modifications to the original model card are in red or green Stable Diffusion is a latent text-to-image diffusion model capable of generating photo-realistic images given any text input. For more information about how Stable Diffusion functions, please have a look at 's Stable Diffusion blog. The Stable-Diffusion-v1-5 checkpoint was initialized with the weights of the Stable-Diffusion-v1-2 checkpoint and subsequently fine-tuned on 595k steps at resolution 512x512 on "laion-aesthetics v2 5+" and 10% dropping of the text-conditioning to improve classifier-free guidance sampling. You can use this both with the Diffusers library and RunwayML GitHub repository ( now deprecated ), ComfyUI… Open weights creativeml-openrail-m 860M parameters diffusers [View model](https://savrn.com/models/stable-diffusion-v1-5) Model · Text to image ### [dreamshaper-7](https://savrn.com/models/dreamshaper-7) [Lykon](https://savrn.com/model-publishers/lykon) lykon/dreamshaper-7 is a Stable Diffusion model that has been fine-tuned on runwayml/stable-diffusion-v1-5. For more general information on how to run text-to-image models with Diffusers, see the docs. - Version 8 focuses on improving what V7 started. Might be harder to do photorealism compared to realism focused models, as it might be hard to do anime compared to anime focused models, but it can do both pretty well if you're skilled enough. Check the examples! - Version 7 improves lora support, NSFW and realism. If you're interested in "absolute" realism, try AbsoluteReality. - Version 6 adds more lora support and more style in general. It should also be better at generating directly at… Open weights creativeml-openrail-m 860M parameters diffusers [View model](https://savrn.com/models/dreamshaper-7) Model · Text to image ### [dreamshaper-8](https://savrn.com/models/dreamshaper-8) [Lykon](https://savrn.com/model-publishers/lykon) lykon/dreamshaper-8 is a Stable Diffusion model that has been fine-tuned on runwayml/stable-diffusion-v1-5. For more general information on how to run text-to-image models with Diffusers, see the docs. - Version 8 focuses on improving what V7 started. Might be harder to do photorealism compared to realism focused models, as it might be hard to do anime compared to anime focused models, but it can do both pretty well if you're skilled enough. Check the examples! - Version 7 improves lora support, NSFW and realism. If you're interested in "absolute" realism, try AbsoluteReality. - Version 6 adds more lora support and more style in general. It should also be better at generating directly at… Open weights creativeml-openrail-m 860M parameters diffusers [View model](https://savrn.com/models/dreamshaper-8) Model · Text to image ### [LCM_Dreamshaper_v7](https://savrn.com/models/lcm-dreamshaper-v7) [Simian Luo](https://savrn.com/model-publishers/simianluo) Distilled from Dreamshaper v7 fine-tune of Stable-Diffusion v1-5 with only 4,000 training iterations (~32 A100 GPU Hours). By distilling classifier-free guidance into the model's input, LCM can generate high-quality images in very short inference time. We compare the inference time at the setting of 768 x 768 resolution, CFG scale w=8, batchsize=4, using a A800 GPU. You can try out Latency Consistency Models directly on: To run the model yourself, you can leverage the Diffusers library: 1. Install the library: 2. Run the model: For more information, please have a look at the official docs: https://huggingface.co/docs/diffusers/api/pipelines/latentconsistencymodels#latent-consistency-models… Open weights mit 860M parameters diffusers [View model](https://savrn.com/models/lcm-dreamshaper-v7) Model · Text to image ### [sd-turbo](https://savrn.com/models/sd-turbo) [Stability AI](https://savrn.com/model-publishers/stabilityai) SD-Turbo is a fast generative text-to-image model that can synthesize photorealistic images from a text prompt in a single network evaluation. We release SD-Turbo as a research artifact, and to study small, distilled text-to-image models. For increased quality and prompt understanding, we recommend SDXL-Turbo. Please note: For commercial use, please refer to https://stability.ai/license. SD-Turbo is a distilled version of Stable Diffusion 2.1, trained for real-time synthesis. SD-Turbo is based on a novel training method called Adversarial Diffusion Distillation (ADD) (see the technical report), which allows sampling large-scale foundational image diffusion models in 1 to 4 steps at high… Open weights 866M parameters diffusers [View model](https://savrn.com/models/sd-turbo) Model · Text to image ### [controlnet-openpose-sdxl-1.0](https://savrn.com/models/controlnet-openpose-sdxl-1-0) [Qi](https://savrn.com/model-publishers/xinsir) thanks feiyuuu for report the problem. When using the default pose line the performance may be unstable, this is because the pose label use more thick line in training to have a better look. This difference can be fix by using the following method: Find the util.py in controlnetaux python package, usually the path is like: /your anaconda3 path/envs/your env name/lib/python3.8/site-packages/controlnetaux/openpose/util.py Replace the drawbodypose function with the following code: Use the code below to get started with the model. HumanArt [https://github.com/IDEA-Research/HumanArt], select 2000 images with ground truth pose annotations to generate images and calculate mAP. We are the SOTA… Open weights apache-2.0 1.3B parameters diffusers [View model](https://savrn.com/models/controlnet-openpose-sdxl-1-0) ## CompVis [All models and datasets](https://savrn.com/model-publishers/compvis) ## Versions - [133a221b8aa7](https://savrn.com/models/stable-diffusion-v1-4/versions/133a221b8aa7) · current 2026-09-18 ## Explore More - [All text to image models](https://savrn.com/models/tasks/text-to-image) - [All models under creativeml-openrail-m](https://savrn.com/models/licenses/creativeml-openrail-m) - [Model comparisons](https://savrn.com/models/comparisons) - [The model directory](https://savrn.com/models) - [Open model prices by host](https://savrn.com/ai-index/pricing/open-models) ## Source - Repository metadata, read 2026-10-19. - [Hugging Face record](https://huggingface.co/CompVis/stable-diffusion-v1-4) - [How the hub is built](https://savrn.com/model-hub/methodology)