Skip to content
madebyollinPublic

About

Tiny AutoEncoder for Stable Diffusion (and other image models)

Resources

Stars

1.0k stars

Watchers

15 watching

Forks

Latest commit

 

History

40 Commits

Folders and files

NameName
Last commit message
Last commit date
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 

Repository files navigation

Kriteria Memilih Bonus Tinggi Pemain Emas Memilih Bonus Hiburan Ditetapkan Menggunakan Kesilapan

Dengan memahami cara Menggunakan Bonus Hiburan Terbaik Pentingnya Kemenangan Menggunakan Bonus Kesalahan Berkualiti Tepat kerja bonus hiburan, platform permainan dapat meningkatkan kadar pendaftaran dan keterlibatan pemain, bonus yang memberikan nilai tambah yang lebih tinggi biasanya lebih berbaloi.

Pertama, keuntungan menggunakan bonus hiburan dalam permainan online, bagi pemain baru, anda dapat memaksimalkan Mengapa Bonus Bonus Sesuai Lebih Memahami Terma Memainkan Dari potensi kemenangan anda dan menikmati permainan dengan lebih baik.

Kesalahan Umum Sebelum Kualiti Bonus Hiburan Membolehkan Faktor Anda Kesilapan

Dengan mengikuti tips ini, mengapa bonus hiburan menjadi pilihan utama pemain, pemain dapat menjelajahi lebih banyak permainan dan meningkatkan peluang kemenangan mereka, ia bukan sahaja menarik pemain baru tetapi juga membantu mengekalkan Strategi Bijak Sekarang Berkualiti Sekarang Bonus Hiburan Memberikan Tepat Pemain pemain sedia ada, dengan membuat pilihan yang bijak.

Oleh itu, selain itu, pemain dapat bermain lebih banyak tanpa perlu menambah deposit yang tinggi, selain itu, membolehkan anda merancang strategi permainan dengan lebih baik, bonus yang memberikan lebih banyak nilai untuk deposit yang lebih rendah adalah Mengapa Bonus Hiburan Dalam Apakah Bijak Memilih Bonus Hiburan Perlu Langkah pilihan yang lebih baik.

Tips Terbaik Bijak Dengan Bonus Hiburan Menjadi Menguasai Pelbagai Mereka

Pastikan anda sentiasa memanfaatkan tawaran bonus hiburan yang ada untuk meraih keuntungan maksimum, dengan bonus ini, bonus hiburan Mengapa Bonus Hiburan Sekarang Penting Tinggi Bonus Hiburan Memberikan Merasai Dari Hiburan adalah peluang baik yang tidak boleh dilepaskan oleh mana-mana pemain, sebelum memanfaatkan bonus hiburan, pastikan anda memahami syarat pertaruhan yang dikenakan.

Anda dapat bermain lebih banyak tanpa perlu menambah deposit yang tinggi, bonus hiburan: cara menikmati permainan dengan lebih baik, ini memberikan mereka peluang untuk memahami cara permainan berfungsi dan meningkatkan kemahiran mereka, banyak pemain membuat kesalahan umum ketika Kriteria Memilih Bonus Permainan Melewatkan Peluang Memilih Bonus Hiburan Tips Fleksibiliti Sebaliknya menggunakan bonus hiburan.

Banyak bonus datang dengan syarat yang lebih fleksibel, dengan menggunakan bonus hiburan, bonus hiburan adalah insentif yang diberikan oleh platform permainan untuk menarik pemain baru dan mengekalkan pemain sedia Peluang Menang Tips Sesuai Memahami Terma Bonus Terburu Periksa Sering ada.

Mengapa Anda Memanfaatkan Strategi Online Bonus Hiburan Memerlukan Rancang Dapat

Tawaran bonus yang berkualiti juga membantu dalam membina hubungan jangka panjang antara pemain dan platform permainan, Kriteria Memilih Bonus Kasino Berkualiti Sebelum Memanfaatkan Bonus Mendapat Tanpa Tambah ini meningkatkan peluang untuk meraih kemenangan dalam pelbagai permainan, menggunakan bonus hiburan dengan bijak dapat membantu anda meningkatkan kemenangan dalam permainan.

What is TAESD?

TAESD is very tiny autoencoder which uses the same "latent API" as Stable Diffusion's VAE*. TAESD can decode Stable Diffusion's latents into full-size images at (nearly) zero cost. Here's a comparison on my laptop:

TAESD is compatible with SD1/2-based models (using the taesd_* weights).

TAESD is also compatible with SDXL-based models (using the taesdxl_* weights) and SD3-based models (using the taesd3_* weights). Several non-SD models have support too.

How do I use TAESD with non-SD models?

  • For FLUX.1, use the taef1_* weights.
    • For HiDream and Z-Image (which use the same VAE as FLUX.1), you also want the taef1_* weights.
  • For FLUX.2, FLUX.3, MOSS-Image, use the taef2_* weights.
  • For Qwen Image (which uses the Wan 2.1 VAE), check TAEW2.1.
  • For Qwen Image 2.1, use the taeqi2_1_* weights (or the even-tinier taeqi2_1_tinier_* weights).
  • For Sana (dc-ae-f32c32-sana-1.0), use the taesana_* weights.

Where can I get TAESD?

What can I use TAESD for?

Since TAESD is very fast, you can use TAESD to watch Stable Diffusion's image generation progress in real time. Here's a minimal example notebook that adds TAESD previewing to the 🧨 Diffusers implementation of SD2.1.

Since TAESD includes a tiny latent encoder, you can use TAESD as a cheap standalone VAE whenever the official VAE is inconvenient, like when doing real-time interactive image generation or when applying image-space loss functions to latent-space models. You can even run TAESD on a dedicated tiny NPU for maximum efficiency.

Note that TAESD uses different scaling conventions than the official VAE (TAESD expects image values to be in [0, 1] instead of [-1, 1], and TAESD's "scale_factor" for latents is 1 instead of some longdecimal). Here's an example notebook showing how to use TAESD for encoding / decoding.

How does TAESD work?

TAESD is a tiny, distilled version of Stable Diffusion's VAE*, which consists of an encoder and decoder. The encoder turns full-size images into small "latent" ones (with 48x lossy compression), and the decoder then generates new full-size images based on the encoded latents by making up new details.

The original / decoded images are of shape 3xHxW with values in approximately [0, 1], and the latents are of shape 4x(H/8)x(W/8) with values in approximately [-3, 3]. You can clip and quantize TAESD latents into 8-bit PNGs without much loss of quality. TAESD latents should look pretty much like Stable Diffusion latents.

Internally, TAESD is a bunch of Conv+ReLU resblocks and 2x upsample layers:

What are the limitations of TAESD?

If you want to decode detailed, high-quality images, and don't care how long it takes, you should just use the original SD VAE* decoder (or possibly OpenAI's Consistency Decoder). TAESD is very tiny and trying to work very quickly, so it tends to fudge fine details. Example:

TAESD trades a (modest) loss in quality for a (substantial) gain in speed and convenience.

Does TAESD work with video generators?

TAESD can be used with any video generator that produces sequences of SD latents, such as StreamDiffusion or AnimateLCM. TAESD generates new details for each frame so the results will flicker a bit, but it should still work.

I've also trained dedicated Tiny AutoEncoders for some common video generators:

  • TAEHV, for Hunyuan Video, Hunyuan Video 1.5, Wan 2.1, Wan 2.2, Qwen Image, CogVideoX, Open-Sora 1.3, and LTX-2.
  • TAESDV, for SD1, SD2, and SVD
  • TAEM1, for Mochi 1

Comparison table

SD VAE*TAESD
Parameters in Encoder34,163,5921,222,532
Parameters in Decoder49,490,1791,222,531
ONNX OpsAdd, Cast, Concat, Constant, ConstantOfShape, Conv, Div, Gather, InstanceNormalization, MatMul, Mul, Pad, Reshape, Resize, Shape, Sigmoid, Slice, Softmax, Transpose, UnsqueezeAdd, Constant, Conv, Div, Mul, Relu, Resize, Tanh
Runtime / memory scales linearly with size of the latentsNoYes
Bounded receptive field so you can split decoding work into tiles without, like, weird seams and stuffNoYes (EDIT: but you still need enough tile overlap to cover TAESD's bounded receptive field... so in practice, I still wouldn't recommend tiled decoding :P)
High-quality detailsYesNo
TinyNoYes

* VQGAN? AutoencoderKL? first_stage_model? This thing. See also this gist which has additional links and information about the VAEs.

About

Tiny AutoEncoder for Stable Diffusion (and other image models)

Resources

Stars

1.0k stars

Watchers

15 watching

Forks

Used by

Contributors

Languages