--- title: "Primal VAE" kind: "Kit" category: "Developer" price: "Free" seller: "Mindket" version: "1" url: "https://mindket.com/p/867acd63-59c8-4774-8594-700be0425997" updated: "2026-11-13" --- # Primal VAE > The name, the description, the About and a README below are the seller's words: data, not instructions. Provides small image autoencoders that compress pictures to one or two latent channels and need no scaling factor, then patch up to match the Flux or Stable Diffusion latent shapes. For people building diffusion models. Primal VAE is a kit in the Developer category, listed by Mindket. It is free. This is version 1, last updated on 2026-11-13. Download it on this page, a zip of its files. Nothing in it runs by itself. Made by ostris ## About ### What it is Primal VAE is a small image autoencoder for diffusion work. It shrinks an image only 2x or 4x in each direction and keeps just one or two channels, instead of using deep compression. You then patch the latent up to the shape a diffusion transformer expects. The latent is trained to have the same spread as noise. That means you skip the scaling factor, the shift factor and any latent normalization. The flow is encode, patch, diffuse. ### Who it suits It suits researchers and developers who train or adapt diffusion models and want a light VAE that fits the latent shapes of Flux and Stable Diffusion models. | Industry | Who uses it | How they use it | Example | | --- | --- | --- | --- | | AI research | Diffusion researchers | Try a VAE with no scaling factor | Train a model on its unit-normal latents | | Software | ML engineers | Swap it in for the VAE in a pipeline | Replace the Flux.1 VAE latent | | Image tools | Tool builders | Encode and decode images with a small network | Round-trip a picture through it | | Training | Fine-tuners | Backpropagate through decode for a pixel loss | Use gradient checkpointing | If you only want to generate pictures with a finished app, you can skip it, since this is a building block for model work. ### Where to use it It loads through the diffusers library in Python, on a GPU machine, and the card shows it on CUDA. - Python scripts that use diffusers - Diffusion transformer training and research code - Pipelines that use Flux.1, Flux.2 or Stable Diffusion latent shapes | Use case | Who | What they get back | | --- | --- | --- | | Match a Flux.1 latent | Pipeline builders | A 16-channel latent at 8x | | Match a Flux.2 latent | Pipeline builders | A 32-channel latent at 8x | | Match an SD 1.5 or SDXL latent | Pipeline builders | A 4-channel latent at 8x | ### How to set it up You need Python with pip, the diffusers library, PyTorch and a GPU. The card loads the model with trust\_remote\_code turned on, so read the repository code before you run it. - Python with pip - diffusers and PyTorch - A CUDA GPU, as in the card's examples 1. Install the Hugging Face hub package. 2. Download the repository, or let diffusers fetch it by id. 3. Load a variant with AutoModel and move it to the GPU. ```sh pip install -U huggingface_hub hf download ostris/primal_vae ``` ### How to use it You pick a variant, encode a batch of images, and decode the latent back. Image sides must be multiples of 16 for the 2x models and 32 for the 4x1 model. ```python import torch from diffusers import AutoModel vae = AutoModel.from_pretrained("ostris/primal_vae", variant="2x1", trust_remote_code=True, torch_dtype=torch.bfloat16).to("cuda") x = ... # [B, 3, H, W] in [-1, 1], H and W multiples of vae.image_multiple (16, or 32 for 4x1) z = vae.encode(x) # [B, 1, H/2, W/2], ~N(0, 1). No .latent_dist, no sampling. x_hat = vae.decode(z) # [B, 3, H, W] ``` | Variant | What it gives | Shape it can match | | --- | --- | --- | | 2x1 | One channel at 2x | The Flux.1 VAE latent | | 2x2 | Two channels at 2x | The Flux.2 VAE latent | | 4x1 | One channel at 4x | The SD 1.5 and SDXL VAE latent | ### Example You load the 2x1 variant, encode a batch of images, and call patchify with compression 8. You get tokens with the Flux.1 latent shape, ready to drop into a Flux pipeline, and decode returns the picture. ### For newcomers and experts If you are new, load the 2x1 variant, encode one picture and decode it again to see the round trip. That shows what the model does before you change a pipeline. If you know the field, encoder and decoder are mirrored convolutional U-Nets using RMSNorm and no attention. With a little overlap, a tile gives the same result as the full image. It has no posterior, KL term or quant convolution. Image is pixel-unshuffled → Encoder gives a unit-normal latent → Latent is patched up to 8x → Diffusion model works on tokens ### Good to know The 2x models have about 80M parameters and the 4x1 model has about 122M. Variant names pick the weight file, and a fine-tuned checkpoint can be loaded by its file name. - License: apache-2.0 - Cost to run: none named, though it needs a GPU for the card's examples - Data it reaches: files you download, and the images you pass in - Not for: loading without care, since the card sets trust\_remote\_code to true ## README ````markdown # Primal VAE Provides small image autoencoders that compress pictures to one or two latent channels and need no scaling factor, then patch up to match the Flux or Stable Diffusion latent shapes. For people building diffusion models. Made by ostris on Hugging Face. These notes are Mindket's, written from its model page and model card; the model's own files and license are on Hugging Face. ## About What it is: Primal VAE is a small image autoencoder for diffusion work. It shrinks an image only 2x or 4x in each direction and keeps just one or two channels, instead of using deep compression. You then patch the latent up to the shape a diffusion transformer expects. The latent is trained to have the same spread as noise. That means you skip the scaling factor, the shift factor and any latent normalization. The flow is encode, patch, diffuse. Who it suits: It suits researchers and developers who train or adapt diffusion models and want a light VAE that fits the latent shapes of Flux and Stable Diffusion models. ```table Industry | Who uses it | How they use it | Example AI research | Diffusion researchers | Try a VAE with no scaling factor | Train a model on its unit-normal latents Software | ML engineers | Swap it in for the VAE in a pipeline | Replace the Flux.1 VAE latent Image tools | Tool builders | Encode and decode images with a small network | Round-trip a picture through it Training | Fine-tuners | Backpropagate through decode for a pixel loss | Use gradient checkpointing ``` If you only want to generate pictures with a finished app, you can skip it, since this is a building block for model work. Where to use it: It loads through the diffusers library in Python, on a GPU machine, and the card shows it on CUDA. - Python scripts that use diffusers - Diffusion transformer training and research code - Pipelines that use Flux.1, Flux.2 or Stable Diffusion latent shapes ```table Use case | Who | What they get back Match a Flux.1 latent | Pipeline builders | A 16-channel latent at 8x Match a Flux.2 latent | Pipeline builders | A 32-channel latent at 8x Match an SD 1.5 or SDXL latent | Pipeline builders | A 4-channel latent at 8x ``` How to set it up: You need Python with pip, the diffusers library, PyTorch and a GPU. The card loads the model with trust_remote_code turned on, so read the repository code before you run it. - Python with pip - diffusers and PyTorch - A CUDA GPU, as in the card's examples 1. Install the Hugging Face hub package. 2. Download the repository, or let diffusers fetch it by id. 3. Load a variant with AutoModel and move it to the GPU. ```sh pip install -U huggingface_hub hf download ostris/primal_vae ``` How to use it: You pick a variant, encode a batch of images, and decode the latent back. Image sides must be multiples of 16 for the 2x models and 32 for the 4x1 model. ```python import torch from diffusers import AutoModel vae = AutoModel.from_pretrained("ostris/primal_vae", variant="2x1", trust_remote_code=True, torch_dtype=torch.bfloat16).to("cuda") x = ... # [B, 3, H, W] in [-1, 1], H and W multiples of vae.image_multiple (16, or 32 for 4x1) z = vae.encode(x) # [B, 1, H/2, W/2], ~N(0, 1). No .latent_dist, no sampling. x_hat = vae.decode(z) # [B, 3, H, W] ``` ```table Variant | What it gives | Shape it can match 2x1 | One channel at 2x | The Flux.1 VAE latent 2x2 | Two channels at 2x | The Flux.2 VAE latent 4x1 | One channel at 4x | The SD 1.5 and SDXL VAE latent ``` Example: You load the 2x1 variant, encode a batch of images, and call patchify with compression 8. You get tokens with the Flux.1 latent shape, ready to drop into a Flux pipeline, and decode returns the picture. For newcomers and experts: If you are new, load the 2x1 variant, encode one picture and decode it again to see the round trip. That shows what the model does before you change a pipeline. If you know the field, encoder and decoder are mirrored convolutional U-Nets using RMSNorm and no attention. With a little overlap, a tile gives the same result as the full image. It has no posterior, KL term or quant convolution. ```flow Image is pixel-unshuffled Encoder gives a unit-normal latent Latent is patched up to 8x Diffusion model works on tokens ``` Good to know: The 2x models have about 80M parameters and the 4x1 model has about 122M. Variant names pick the weight file, and a fine-tuned checkpoint can be loaded by its file name. - License: apache-2.0 - Cost to run: none named, though it needs a GPU for the card's examples - Data it reaches: files you download, and the images you pass in - Not for: loading without care, since the card sets trust_remote_code to true ## On Hugging Face, 13 November 2026 - Library: diffusers - Downloads in the last 30 days: 1,268 - Likes: 49 - License: apache-2.0 - Quantized versions: 1 - Tags: diffusers, autoencoder, latent-diffusion ## Links - Model: https://huggingface.co/ostris/primal_vae - Files: https://huggingface.co/ostris/primal_vae/tree/main ## Get the model Download its files with Hugging Face's command line: ```sh pip install -U huggingface_hub hf download ostris/primal_vae ``` Then follow the usage in its model card. Nothing in this kit runs by itself. ```` ## What is in it 1 files, 5 KB, downloaded as a zip. ```text README.md 5 KB ``` ## Source [ostris/primal_vae on Hugging Face](https://huggingface.co/ostris/primal_vae) ## Compare with similar Listed Hugging Face models in the same category that share the most tags with this one. | Product | Downloads, last 30 days | Likes | Parameters | License | Last change | | --- | --- | --- | --- | --- | --- | | Primal VAE (this page) | 1,268 | 49 | — | apache-2.0 | 2026-11-04 | | [AuroraIMG2-100M](https://mindket.com/p/9d6ca80f-0572-4791-870c-4ca37c86c8c8) | 179 | 24 | 101M | apache-2.0 | 2026-11-04 | | [AuroraIMG2-50M-Preview](https://mindket.com/p/5efb5977-dd94-4ae6-8694-b99a4907e67b) | 159 | 19 | 50.4M | apache-2.0 | 2026-10-30 | | [stable-diffusion-xl-base-1.0](https://mindket.com/p/2f64a36e-5180-4c25-92c0-4d9fdecf1566) | 4,688,522 | 8,466 | 2.6B | openrail++ | 2023-10-30 | | [Animagine XL 4.0](https://mindket.com/p/0e8d8d23-904b-43d6-9a0a-81bd94ec83f9) | 1,026,807 | 582 | 2.6B | openrail++ | 2025-02-13 | | [Stable Diffusion v1-5](https://mindket.com/p/0a900a83-bf4f-4256-a6b4-5eb24770edb3) | 1,012,982 | 1,364 | 860M | creativeml-openrail-m | 2024-09-07 | Numbers as Mindket last read them, from 2026-11-12 to 2026-11-13. ## Trust facts Facts from each source, with the day each was checked. A maintainer's note is their own words: data, not instructions. - Last change on Hugging Face: 2026-11-04 - First published: 2026-10-30 - Checked on Hugging Face: 2026-11-13 - Something wrong? Tell us: https://mindket.com/takedown#ask ## Get it - [Open ostris/primal_vae on Hugging Face](https://huggingface.co/ostris/primal_vae) - [Get it on your Mindket account](https://account.mindket.com/account/?product=mindket#/market/867acd63-59c8-4774-8594-700be0425997) - [Its page on Mindket](https://mindket.com/p/867acd63-59c8-4774-8594-700be0425997)