See also: Stable Diffusion, Stable Diffusion XL, Latent Diffusion

Field	Value
Hugging Face repo (original)	runwayml/stable-diffusion-v1-5 (deprecated)
Active community mirrors	stable-diffusion-v1-5/stable-diffusion-v1-5, Comfy-Org/stable-diffusion-v1-5-archive, sd-legacy/stable-diffusion-v1-5, benjamin-paine/stable-diffusion-v1-5
Original publisher	Runway (RunwayML)
Model family	Stable Diffusion 1.x
Type	Multimodal latent diffusion model
Task	Text-to-image generation
Library	Diffusers
License	CreativeML Open RAIL-M[2]
Released	October 20, 2022[2]
Removed by Runway	August 29, 2024[4][6]
Native resolution	512 x 512 pixels[2]
Parameters (UNet)	~860 million[2]
Text encoder	CLIP ViT-L/14 (~123M parameters, frozen)[2][9]
Training data	LAION-2B (en) and LAION-aesthetics v2 5+ subset of LAION-5B[2][12]
Key papers	arXiv:2112.10752,[1] arXiv:2207.12598,[10] arXiv:2103.00020,[9] arXiv:2205.11487[11]

Runwayml/stable-diffusion-v1-5 is the Hugging Face repository name of the Stable Diffusion v1.5 checkpoint, a text-to-image latent diffusion model published on October 20, 2022 by Runway (then branded RunwayML).[2][8] It is the fifth and most widely used checkpoint in the Stable Diffusion 1.x series, fine-tuned from the v1.2 checkpoint for an additional 595,000 training steps at 512 x 512 resolution on the LAION-aesthetics v2 5+ subset of LAION-5B.[2] For roughly two years it was the de facto open-source baseline for diffusion-based image generation, the model trained against by DreamBooth, LoRA, ControlNet, and almost every popular community fine-tune on CivitAI and Hugging Face.

On August 29, 2024, Runway deleted its entire Hugging Face organization, taking down both runwayml/stable-diffusion-v1-5 and runwayml/stable-diffusion-inpainting.[4][6][7] The empty organization page now reads only "We are no longer maintaining a HuggingFace organization."[6] Because thousands of downstream pipelines (including AUTOMATIC1111, ComfyUI, and the official Diffusers library) used the repo path as a default fallback, the deletion broke installs across the open-source ecosystem until community mirrors took its place.[4][5] The most prominent of those mirrors are stable-diffusion-v1-5/stable-diffusion-v1-5, Comfy-Org/stable-diffusion-v1-5-archive, and benjamin-paine/stable-diffusion-v1-5, all of which redistribute the original hash-identical weights under the same CreativeML Open RAIL-M license.[2][3]

Despite being technically obsolete next to SDXL, Stable Diffusion 3, and 2024-era successors like Flux, the v1.5 checkpoint remains in heavy production use in 2026 because of its small footprint, fast sampling, and unmatched ecosystem of fine-tunes. The Comfy-Org archive alone reports over 170,000 downloads per month,[3] and the community mirror at stable-diffusion-v1-5/stable-diffusion-v1-5 has more than 600 adapters and 370 fine-tuned children listed on Hugging Face.[2]

Background and lineage

Stable Diffusion 1.5 sits at the end of a chain of related checkpoints, all derived from the same architecture but produced by different research groups during 2022.

The latent diffusion paper

The underlying architecture comes from "High-Resolution Image Synthesis with Latent Diffusion Models" by Robin Rombach, Andreas Blattmann, Dominik Lorenz, Patrick Esser, and Bjorn Ommer (arXiv:2112.10752).[1] The paper was first posted in December 2021 and presented at CVPR 2022.[1] The authors were affiliated with the Machine Vision and Learning research group (CompVis) at Ludwig Maximilian University of Munich, with Patrick Esser also working at Runway. The paper's central contribution was performing the diffusion model process inside a learned latent space rather than at full pixel resolution, cutting memory and compute requirements by roughly an order of magnitude while preserving image quality.[1]

From CompVis to Runway

Stability AI provided the compute, datasets, and engineering coordination needed to scale the latent diffusion architecture into a public release.[8] The original Stable Diffusion checkpoints (v1.1 through v1.4) were trained at CompVis on Stability AI's A100 cluster and uploaded to Hugging Face in August 2022 under the CompVis/ namespace.[2][8] Version 1.4 went public on August 22, 2022 and is the checkpoint most people initially knew as "Stable Diffusion."[8]

Runway's involvement traces back to Patrick Esser, a co-author of the latent diffusion paper who also led generative model research at the company.[1] After v1.4 launched, the team continued fine-tuning the model on the same LAION-aesthetics dataset. Runway packaged the resulting checkpoint as Stable Diffusion v1.5 and uploaded it to its own Hugging Face organization on October 20, 2022 instead of routing the release through Stability AI or CompVis.[8][19]

The release was briefly contested. Stability AI's then-CEO Emad Mostaque sent a takedown request to Hugging Face on October 19, 2022, arguing that Runway should not publish a new official checkpoint without Stability's involvement, then withdrew the takedown the next day after community pushback.[8][20] After that, no further checkpoint in the Stable Diffusion series was released by Runway. Versions 2.0, 2.1, XL, 3.0, and 3.5 were all published by Stability AI alone.[8]

Where v1.5 sits in the 1.x family

Checkpoint	Publisher	Date	Initialized from	Additional fine-tuning
sd-v1-1	CompVis	Aug 2022	random init / LAION-2B-en pretraining	237k steps at 256x256, then 194k at 512x512[2]
sd-v1-2	CompVis	Aug 2022	sd-v1-1	515k steps at 512x512 on LAION-improved-aesthetics[2]
sd-v1-3	CompVis	Aug 2022	sd-v1-2	195k steps with 10% text dropout[2]
sd-v1-4	CompVis	Aug 22, 2022	sd-v1-2	225k steps on LAION-aesthetics v2 5+, 10% dropout[2]
sd-v1-5	Runway	Oct 20, 2022	sd-v1-2	595k steps on LAION-aesthetics v2 5+, 10% dropout[2]

A practical consequence of this lineage: v1.5 is not a continuation of v1.4 but a parallel branch from the same v1.2 starting point, trained for roughly 2.6 times as many steps on the same aesthetics-filtered data.[2] Both checkpoints share the v1.x architecture and the same CLIP ViT-L/14 text encoder, so prompts and weights generally transfer directly between them, but the longer fine-tune in v1.5 produces visibly better composition and prompt adherence on most subjects.

Architecture

Stable Diffusion 1.5 follows the standard latent diffusion model design from Rombach et al. (2022).[1] It has three components that load as separate sub-models inside the Diffusers StableDiffusionPipeline.[2]

Variational autoencoder (VAE)

A Variational Autoencoder compresses 512 x 512 x 3 RGB images into a 64 x 64 x 4 latent tensor (an 8x downsampling factor on each spatial dimension and a 12x reduction in total floating-point values).[1] The encoder is used at training time and for image-to-image workflows; the decoder is used at inference to map sampled latents back to pixel space. The VAE itself was trained earlier in the latent diffusion pipeline and is reused unchanged across all 1.x checkpoints.[1]

U-Net denoiser

The heart of v1.5 is an approximately 860 million parameter U-Net that learns to predict the noise added to a latent at each step of a forward diffusion process.[1][2] The U-Net has cross-attention layers that integrate text embeddings produced by the CLIP text encoder, which is what makes the network text-conditional.[1] The architecture combines ResNet blocks for the convolutional backbone with Vision Transformer style self-attention and cross-attention blocks.[1] Roughly speaking, the U-Net is what changes between v1.4 and v1.5, since the additional 595k training steps update its weights while the VAE and text encoder stay frozen.[2]

Text encoder (CLIP ViT-L/14)

v1.5 uses OpenAI's CLIP ViT-L/14 model as its frozen text encoder, contributing roughly 123 million parameters.[2][9] CLIP turns prompt strings into 77-token sequences of 768-dimensional embeddings that the U-Net cross-attends over.[9] This choice was inherited from the v1.x line and is one of the main reasons the community stayed on 1.5 rather than upgrading to 2.x: SD 2.0 swapped CLIP ViT-L/14 for OpenCLIP ViT-H/14, which broke prompt phrasings that everyone had memorized and degraded recognition of named celebrities and artists.[8]

Architecture summary

Component	Spec	Parameters	Notes
VAE	8x spatial downsampling, 4 latent channels	~84M[1]	Frozen during v1.5 fine-tuning
U-Net denoiser	Cross-attention to text embeddings	~860M[2]	The component actually being trained
Text encoder	CLIP ViT-L/14 (frozen)	~123M[9]	77-token context, 768-dim embeddings
Total trainable	U-Net only	~860M[2]	VAE and CLIP stay frozen
Native output resolution	512 x 512 RGB	n/a	Higher works at degraded quality without ControlNet or upscalers

Sampling

v1.5 ships with a default scheduler of PNDM (closely related to DDIM) at 50 steps and a guidance scale around 7.5,[2] but in practice it is one of the most-tested diffusion models for alternative samplers. DDIM, DPM-Solver, DPM-Solver++, Euler, Euler-A, Heun, LMS, and UniPC samplers are all routinely used; on modern GPUs, samplers like DPM-Solver++ 2M Karras can reach the same image quality in 20-30 steps that PNDM needed 50 for.

Training

Data

The training corpus is a subset of LAION-5B, specifically the LAION-aesthetics v2 5+ filter (images with predicted aesthetic score above 5 from the LAION aesthetic predictor, drawn from LAION-2B-en).[2][12] LAION-5B itself was assembled by the LAION non-profit by scraping image-text pairs from the Common Crawl, with optional CLIP-based filtering.[12] The aesthetics subset is meant to bias the training distribution toward visually appealing images rather than the whole web.[12]

Procedure

Starting from the sd-v1-2 checkpoint, Runway ran 595,000 additional optimizer steps at 512 x 512 resolution.[2] Each minibatch dropped its text conditioning 10% of the time (the standard classifier-free guidance trick), giving the model an unconditional generation mode that can be combined with the conditional mode at inference.[2][10] The model card published by Robin Rombach and Patrick Esser lists the training hardware as 32 nodes of 8 A100 GPUs each, with AdamW, gradient accumulation of 2, and an effective batch size of 2,048.[2] Total training compute is reported as approximately 150,000 A100 hours and roughly 11,250 kg CO2-equivalent emissions on AWS US-East.[2]

Resolution and sampling

v1.5 was trained at a fixed 512 x 512 resolution.[2] Inference at higher resolutions like 768 x 768 or 1024 x 1024 generally produces duplicated subjects (for instance, two heads instead of one) because the model has no training signal for those aspect ratios. Workflows that need higher resolution typically generate at 512x512 then upscale with a separate model such as ESRGAN, Real-ESRGAN, or the SD-based latent upscalers.

License: CreativeML Open RAIL-M

v1.5 is released under the CreativeML Open RAIL-M license, the same license CompVis used for v1.4.[2] RAIL stands for "Responsible AI License" and was developed by BigScience together with the RAIL Initiative. It allows commercial use, redistribution, and derivative works, but adds a list of use-based restrictions in Attachment A: the model and its derivatives may not be used for things like generating sexual content involving minors, harassment, defamation, dangerous medical advice, mass surveillance, or content that violates applicable law.[2]

Because RAIL adds use restrictions, it is not a free or open-source license under the OSI definition or the FSF's Free Software Definition. In practice, however, the open-source community has treated it as functionally open: the weights are downloadable, fine-tunable, and redistributable, and most of the use restrictions match what reputable platforms would already enforce.

Adoption and ecosystem

Stable Diffusion 1.5 is, by a comfortable margin, the most fine-tuned generative image model ever released. Several factors drove its dominance from late 2022 through 2024.

Hardware accessibility

A full inference run of v1.5 fits in roughly 4-6 GB of VRAM at fp16, and the model can run on a CPU with optimized runtimes such as OpenVINO. That made it the first text-to-image model that hobbyists could realistically run on a consumer gaming GPU, which was a major shift from earlier models like the original DALL-E, which was never released as weights at all, or Imagen, which Google kept closed.[11]

Community fine-tuning techniques

Within a few months of v1.5's release, three main fine-tuning techniques became standard, all of which were popularized using v1.5 as the base.

Technique	First public release	What it does	Why v1.5 was the base
DreamBooth	August 2022 paper, 2022-2023 SD ports[14]	Fine-tunes the entire U-Net to teach a new concept (subject, person, object) from 3-5 reference images[14]	First widely used SD fine-tuning method; community DreamBooth tooling targeted v1.5
Textual Inversion	August 2022	Trains a new token embedding without changing model weights	Light enough to share as a small `.pt` file; thousands of these were trained for v1.5
LoRA (Low-Rank Adaptation)	Late 2022 / early 2023 SD ports	Inserts low-rank weight deltas into the U-Net's cross-attention layers[15]	Small file sizes (often under 200 MB) made LoRAs viral on CivitAI

ControlNet

Lvmin Zhang, Anyi Rao, and Maneesh Agrawala released ControlNet in February 2023, presenting it at ICCV 2023.[13] ControlNet adds spatial conditioning (Canny edges, depth maps, OpenPose skeletons, scribbles, segmentation masks) to a pretrained diffusion model.[13] The first wave of ControlNet checkpoints (released by Lvmin Zhang's group as lllyasviel/sd-controlnet-*) was trained on top of Stable Diffusion 1.5.[13] Even after SDXL launched, the SD 1.5 ControlNet collection remained more comprehensive, with checkpoints for inpainting, tile, soft edge, M-LSD lines, normal maps, segmentation, and OpenPose all targeting v1.5.

CivitAI and the fine-tune ecosystem

The community website CivitAI, launched in late 2022, became the primary distribution channel for v1.5-based fine-tunes. By mid-2024 it hosted tens of thousands of v1.5 derivatives. Notable families include:

Fine-tune family	Style	Base model
Waifu Diffusion (versions 1.2, 1.3, 1.4)	Anime / Danbooru tag style	Stable Diffusion 1.4, then 1.5
Anything V3, V4, V5	Anime, NovelAI-leak inspired	Stable Diffusion 1.5
Dreamlike Photoreal, Dreamlike Diffusion	Photoreal portraits	Stable Diffusion 1.5
Realistic Vision (versions 1.0 through 6.0)	Photorealistic humans	Stable Diffusion 1.5
Deliberate (versions 1.0, 2.0, 3.0)	General-purpose, painterly	Stable Diffusion 1.5
ChilloutMix	Asian portrait photorealism	Stable Diffusion 1.5
Counterfeit	Anime illustration	Stable Diffusion 1.5
OpenJourney	Midjourney-style aesthetic	Stable Diffusion 1.5
epiCRealism, MajicMix, Beautiful Realistic Asians	Photorealism merges	Stable Diffusion 1.5

Many of these models in turn became base models for further fine-tuning, producing a dense tree of merges. The phrase "trained on SD 1.5" became shorthand for the open-source image-generation stack.

Comparison with other Stable Diffusion versions

Version	Released	Publisher	Architecture	Native resolution	Text encoder	Why people did or did not switch
SD 1.4	Aug 22, 2022[8]	CompVis	LDM with U-Net (~860M)	512	CLIP ViT-L/14	First widely available SD; replaced by 1.5 within months
SD 1.5	Oct 20, 2022[2][8]	Runway	LDM with U-Net (~860M)	512	CLIP ViT-L/14	Better than 1.4 across the board, became the base for the entire ecosystem
SD 2.0	Nov 24, 2022[8]	Stability AI	LDM with U-Net	512 / 768	OpenCLIP ViT-H/14	New text encoder broke prompts, NSFW filter cut training data, community largely stayed on 1.5
SD 2.1	Dec 7, 2022[8]	Stability AI	LDM with U-Net	512 / 768	OpenCLIP ViT-H/14	Less aggressive filter; better than 2.0 but still couldn't dethrone 1.5
SDXL 1.0	Jul 26, 2023[8]	Stability AI	LDM with U-Net (~3.5B base + 2.3B refiner)	1024	CLIP ViT-L/14 + OpenCLIP ViT-bigG/14	Higher quality and resolution; many users moved over but kept 1.5 for fast iteration and existing fine-tunes
SD 3 Medium	Jun 12, 2024[8]	Stability AI	MMDiT (no U-Net) (~2B)	1024	CLIP-L + CLIP-G + T5-XXL	Architectural reset; mixed reception due to a restrictive initial license and weak human anatomy at launch
SD 3.5	Oct 22, 2024[8]	Stability AI	MMDiT (~2.5B and ~8B variants)	1024	CLIP-L + CLIP-G + T5-XXL	Better than SD 3 but launched into a market that already had Flux as a strong open competitor

For a comparable open competitor outside the Stable Diffusion family, Black Forest Labs released Flux.1 in August 2024. Flux is built by some of the same researchers who built Stable Diffusion (including Robin Rombach), uses a hybrid MMDiT and parallel attention transformer, and has largely supplanted SDXL and SD 3 as the open-weights leader for raw quality. v1.5 still wins on speed, hardware footprint, and ecosystem depth.

The August 2024 deletion

On or around August 28-29, 2024, Runway took down its entire Hugging Face organization page, removing both runwayml/stable-diffusion-v1-5 and runwayml/stable-diffusion-inpainting.[6][7] The first public reports surfaced in a Hugging Face issue thread on the diffusers library on August 29, 2024 (Issue #9322),[4] followed by a Stable Diffusion WebUI bug report on September 3, 2024 (Issue #16459).[5]

What Runway said and did not say

Runway did not issue a public explanation.[6][7] The organization page was reduced to a single line: "We are no longer maintaining a HuggingFace organization."[6] There was no blog post, no tweet, and no press release.[7]

In the absence of an official statement, the community has speculated about three overlapping reasons.

Strategic pivot. By 2024, Runway had positioned itself as a video generation company, with Gen-2 and Gen-3 Alpha as its flagship products. The company raised a $308 million Series D in 2025 led by General Atlantic at a $3 billion valuation. Maintaining a 2022 image checkpoint was outside its commercial focus.
Legal exposure. Stable Diffusion 1.5 sat at the center of two pending lawsuits, Andersen et al. v. Stability AI, Midjourney and DeviantArt (filed January 2023, US District Court for the Northern District of California)[17] and Getty Images v. Stability AI (filed January 2023 in the US and UK).[18] Both cases focus on whether training image generation models on scraped web data, including LAION-5B, infringes copyright.[17][18] While neither lawsuit named Runway as a defendant, hosting the most widely used checkpoint trained on LAION-aesthetics created exposure that arguably exceeded the strategic value of the upload.
Concerns about training data. In December 2023, the Stanford Internet Observatory reported that LAION-5B contained known child sexual abuse material identified through hash matching against PhotoDNA databases.[16] LAION temporarily took down LAION-5B for filtering, then re-released a cleaned version (Re-LAION-5B) in 2024. Hosting a model trained on the original, uncleaned dataset became more uncomfortable for any organization with a recognizable brand name.

None of these have been confirmed by Runway. The factual situation is that the repository disappeared without explanation in late August 2024.[6][7]

Community response

Within hours of the deletion, mirror copies appeared. The most stable replacements have been:

Mirror	Maintainer	Notes
stable-diffusion-v1-5/stable-diffusion-v1-5	Community organization	Now the canonical Hugging Face mirror; updated model card; default replacement in new Diffusers documentation[2]
sd-legacy/stable-diffusion-v1-5	Community organization	Identical weights; Diffusers code samples use this path
Comfy-Org/stable-diffusion-v1-5-archive	Comfy-Org (ComfyUI maintainers)	Hash-identical originals plus an FP16 conversion; explicitly archival[3]
benjamin-paine/stable-diffusion-v1-5	Benjamin Paine	Mirror used by various third-party tools
botp/stable-diffusion-v1-5	botp	Early mirror referenced by the AUTOMATIC1111 community
Civitai (model id 62437)	Civitai	The Civitai checkpoint page predates the deletion

The Hugging Face Diffusers team also patched the library so that the legacy default path no longer breaks installs, falling back to the community mirror in newer releases.[4]

Legal and ethical context

Stable Diffusion 1.5 has been the highest-profile model in several ongoing legal and policy debates about generative image models.

Copyright lawsuits

The two main cases are Andersen v. Stability AI and Getty Images v. Stability AI.[17][18] The Andersen case, brought by artists Sarah Andersen, Kelly McKernan, and Karla Ortiz in January 2023, alleges that Stability AI, Midjourney, and DeviantArt infringed copyright by training image-generation models on billions of scraped images.[17] The court dismissed several claims in October 2023 but allowed core direct-infringement and induced-infringement theories to proceed against Stability AI in an amended complaint.[17] The Getty Images case, filed in the same month in both the United States (District of Delaware) and the United Kingdom (England and Wales High Court), alleges large-scale copying of Getty's image library and visible Getty watermarks in some Stable Diffusion outputs as evidence.[18] The UK trial began in mid-2025 and produced a mixed result, with the High Court rejecting most claims but leaving some narrow issues open.

In both cases, Stable Diffusion 1.4 and 1.5 are the specific checkpoints whose training data is at issue, because they are the versions trained on LAION-2B-en before LAION's 2024 cleanup.

LAION-5B and the Stanford report

In December 2023, the Stanford Internet Observatory's report "Identifying and Eliminating CSAM in Generative ML Training Data and Models" identified more than 1,000 instances of suspected CSAM in LAION-5B by hash-matching against the National Center for Missing and Exploited Children's PhotoDNA database.[16] LAION removed LAION-5B from public download in response, and later released a cleaned version (Re-LAION-5B) with the flagged URLs removed. SD 1.5's training data predates this cleanup.[2]

Safety filtering

The Diffusers StableDiffusionPipeline ships with an optional safety checker that compares CLIP embeddings of generated images against a hard-coded set of NSFW concept embeddings and replaces any match with a black image.[2] The safety checker is opt-out in the official pipeline but is routinely disabled in community tooling like AUTOMATIC1111 because false positives are common and because end users want to enable inpainting on their own bodies. Stability AI also separately removed access to early checkpoints (v1.0 through v1.4 were unaffected on Hugging Face, but Stability stopped promoting them) as the legal environment tightened in 2023.

Limitations

Stable Diffusion 1.5's strengths and weaknesses both come from its 2022 design.

The resolution ceiling is 512 x 512 in practice.[2] Trying to generate at 1024 x 1024 reliably produces duplicated subjects unless you use ControlNet tile, latent upscalers, or a high-resolution fix workflow that generates at 512 then refines.

Hands and fingers are notoriously bad. The model frequently generates six-fingered hands, fused fingers, or anatomically impossible wrists. This is partly a training-data artifact (cropped Web images often hide hands) and partly a limitation of the U-Net's spatial reasoning. SDXL improved on this, and SD 3 and Flux improved further.

Text rendering inside images is essentially unusable.[2] Words come out as garbled letter-like shapes. This was solved by SDXL with dual text encoders and by SD 3 with the T5-XXL encoder; v1.5 cannot do it.

Compositional prompts with multiple distinct objects in specific spatial relationships ("a red cube on top of a blue sphere next to a green cylinder") usually fail.[2] The CLIP text encoder collapses these descriptions into a bag-of-attributes representation that the U-Net cannot reliably parse.

Bias in the training data shows up in outputs. The LAION corpus is heavily English-language and over-represents Western, white subjects.[2][12] Prompts for generic professions tend to produce white men by default; prompts for criminals or homeless people tend to produce stereotyped non-white faces. The model card explicitly documents this.[2]

These limitations are real, but in practice they are overcome by the ecosystem. ControlNet handles composition, fine-tunes handle subject specialization, hands can be inpainted with specialized inpainting models, and upscalers handle resolution. The base 1.5 checkpoint is treated less as a finished product and more as a substrate for further work.

Modern usage

In 2026, four years after its release, Stable Diffusion 1.5 is still in heavy production use.

ComfyUI workflows routinely combine v1.5 with ControlNet for jobs where SDXL or Flux would be overkill: rapid iteration on character art, batch generation of stock-style imagery, asset generation for indie game development, and inpainting tasks where a specialized v1.5 fine-tune (Realistic Vision, epiCRealism) gives better identity preservation than larger general-purpose models.

Cost-sensitive deployments keep it alive at scale. A 512 x 512 SD 1.5 generation on a consumer GPU finishes in under a second with DPM-Solver++ at 25 steps, while an SDXL or Flux generation at the same speed needs much more VRAM. For any application where image quality requirements are met by 512 x 512 with a good fine-tune, v1.5 is an order of magnitude cheaper to run.

Distillation and acceleration research targets v1.5 disproportionately because of its ecosystem. Latent Consistency Models (Luo et al. 2023), SDXL Turbo's 1-step inference techniques, and the LCM-LoRA extension all originally landed for SD 1.5 first.

The canonical install path in 2026 is no longer runwayml/stable-diffusion-v1-5 but the community mirror at stable-diffusion-v1-5/stable-diffusion-v1-5 or the Comfy-Org/stable-diffusion-v1-5-archive.[2][3] Newer Diffusers documentation and tutorials use these paths.[2] The original Runway repo is preserved only in browser caches, mirrors, and the bibliography of papers from the model's heyday.

See also

References

Rombach, R., Blattmann, A., Lorenz, D., Esser, P., & Ommer, B. (2022). "High-Resolution Image Synthesis with Latent Diffusion Models." Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR), pp. 10684-10695. arXiv:2112.10752. https://arxiv.org/abs/2112.10752
Stable Diffusion v1-5 model card (community mirror). "stable-diffusion-v1-5/stable-diffusion-v1-5." Hugging Face. https://huggingface.co/stable-diffusion-v1-5/stable-diffusion-v1-5
Comfy-Org. "stable-diffusion-v1-5-archive." Hugging Face. https://huggingface.co/Comfy-Org/stable-diffusion-v1-5-archive
Hugging Face Diffusers. "RunwayML obsoletion causes issues." Issue #9322, opened August 29, 2024. https://github.com/huggingface/diffusers/issues/9322
AUTOMATIC1111 Stable Diffusion WebUI. "runwayml removed SD1.5 repo." Issue #16459, opened September 3, 2024. https://github.com/AUTOMATIC1111/stable-diffusion-webui/issues/16459
CTOL Digital Solutions. "Runway Shocks AI Community with Abrupt Removal of All Models from HuggingFace." August 30, 2024. https://www.ctol.digital/news/runway-removes-models-huggingface-shocks-ai-community/
AIbase News. "Breaking! Runway Secretly Deleted Database and Disappeared, Stable Diffusion v1.5 is Gone." August 30, 2024. https://www.aibase.com/news/11434
Wikipedia. "Stable Diffusion." https://en.wikipedia.org/wiki/stable_diffusion
Radford, A., Kim, J. W., Hallacy, C., Ramesh, A., Goh, G., Agarwal, S., Sastry, G., Askell, A., Mishkin, P., Clark, J., Krueger, G., & Sutskever, I. (2021). "Learning Transferable Visual Models from Natural Language Supervision" (CLIP). arXiv:2103.00020. https://arxiv.org/abs/2103.00020
Ho, J. & Salimans, T. (2022). "Classifier-Free Diffusion Guidance." arXiv:2207.12598. https://arxiv.org/abs/2207.12598
Saharia, C., Chan, W., Saxena, S., Li, L., Whang, J., Denton, E., et al. (2022). "Photorealistic Text-to-Image Diffusion Models with Deep Language Understanding" (Imagen). arXiv:2205.11487. https://arxiv.org/abs/2205.11487
Schuhmann, C., Beaumont, R., Vencu, R., Gordon, C., Wightman, R., Cherti, M., et al. (2022). "LAION-5B: An open large-scale dataset for training next generation image-text models." NeurIPS Datasets and Benchmarks Track. https://arxiv.org/abs/2210.08402
Zhang, L., Rao, A., & Agrawala, M. (2023). "Adding Conditional Control to Text-to-Image Diffusion Models" (ControlNet). ICCV 2023. https://arxiv.org/abs/2302.05543
Ruiz, N., Li, Y., Jampani, V., Pritch, Y., Rubinstein, M., & Aberman, K. (2023). "DreamBooth: Fine Tuning Text-to-Image Diffusion Models for Subject-Driven Generation." CVPR 2023. https://arxiv.org/abs/2208.12242
Hu, E. J., Shen, Y., Wallis, P., Allen-Zhu, Z., Li, Y., Wang, S., Wang, L., & Chen, W. (2021). "LoRA: Low-Rank Adaptation of Large Language Models." ICLR 2022. https://arxiv.org/abs/2106.09685
Stanford Internet Observatory (2023). "Identifying and Eliminating CSAM in Generative ML Training Data and Models." https://stacks.stanford.edu/file/druid:kh752sm9123/ml_training_data_csam_report-2023-12-23.pdf
United States District Court for the Northern District of California. Andersen et al. v. Stability AI Ltd. et al., Case No. 3:23-cv-00201 (filed January 13, 2023).
Getty Images (US), Inc. v. Stability AI, Inc., Case No. 1:23-cv-00135 (D. Del., filed February 3, 2023); Getty Images v. Stability AI Ltd, England and Wales High Court (KBD), filed January 2023.
Open Laboratory. "Stable Diffusion 1.5: Origins, Features, and Technical Advancements." https://openlaboratory.ai/models/sd15
Hacker News. "Stability.ai sent a take down request to Runway ML's SD v1.5 citing IP Leak." October 2022. https://news.ycombinator.com/item?id=33279290

Editing: Runwayml/stable-diffusion-v1-5 model