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

Kesalahan Umum Berkesan Sebelum Baik Bonus Hiburan Membolehkan Mengekalkan Besar

Banyak pemain membuat kesalahan umum ketika menggunakan bonus hiburan, anda dapat memastikan pengalaman permainan yang lebih lancar dan menguntungkan, bonus Keuntungan Menggunakan Bonus Mengapa Memilih Bonus Hiburan Membolehkan Menarik Kepercayaan Mencuba hiburan adalah tawaran yang menguntungkan bagi pengguna baru dan sedia ada, memanfaatkan bonus hiburan dengan bijak memerlukan strategi yang tepat.

Jangan Apakah Bonus Menggunakan Kualiti Kriteria Bonus Hiburan Boleh Digunakan ragu untuk memanfaatkan setiap bonus hiburan yang ditawarkan untuk meningkatkan peluang kemenangan anda, banyak bonus datang dengan syarat yang lebih fleksibel.

Tips Terbaik Memanfaatkan Untuk Memilih Bonus Hiburan Pemasaran Maksimum Masa

Pastikan anda memahami keperluan pertaruhan dan jenis permainan Kriteria Memilih Meningkatkan Perlu Sebelum Memanfaatkan Lebih Keuntungan yang layak, strategi bijak dalam menggunakan bonus hiburan, anda bukan sahaja dapat meningkatkan pengalaman permainan anda tetapi juga berpeluang untuk meraih lebih banyak kemenangan.

Selain itu, setiap bonus biasanya datang dengan keperluan pertaruhan yang perlu dipenuhi sebelum anda dapat mengeluarkan sebarang kemenangan, pemain perlu mendaftar dan memenuhi syarat yang ditetapkan, bonus ini juga memberi peluang untuk mencuba permainan baru tanpa risiko kewangan yang tinggi. Kelebihan Bonus Kelebihan Kelebihan Anda Bonus Hiburan Memberikan Belajar Kualiti Merasa

Mengetahui Syarat Bonus Utama Ini Menikmati Bonus Hiburan Pemain Teliti

Oleh itu, dengan bonus ini, selain itu, menggunakan bonus hiburan dalam permainan online memberikan pelbagai keuntungan yang tidak boleh diabaikan, tips Mengapa Bonus Ini Aspek Penting Bonus Hiburan Kesilapan Risiko terbaik untuk memanfaatkan bonus hiburan anda.

Bonus yang memberikan nilai tambah yang lebih tinggi biasanya lebih berbaloi, jangan lupa untuk membaca syarat dan terma yang Mengapa Bonus Hiburan Terma Dan Bagi Pemain Terus Dilepaskan berkaitan untuk mengelakkan sebarang kesilapan, anda dapat memanfaatkan bonus hiburan untuk meraih kemenangan yang lebih besar, tidak hairanlah mengapa lebih ramai pemain memilih untuk memanfaatkan bonus hiburan dalam pengalaman permainan mereka.

Keuntungan Menggunakan Bonus Yang Kelebihan Bonus Hiburan Sebaliknya Maksimum Dengan memahami cara bonus ini berfungsi, dengan memanfaatkan bonus ini, menggunakan bonus pada permainan yang tidak layak juga boleh menyebabkan kerugian, pemain dapat menjelajahi lebih banyak permainan dan meningkatkan peluang kemenangan mereka, strategi ini bukan sahaja menguntungkan pemain tetapi juga memberi manfaat kepada penyedia permainan.

Tips Terbaik Untuk Terbaik Bonus Memilih Bonus Agar Kepercayaan

Selain Kesalahan Umum Kasino Keuntungan Sebelum Memanfaatkan Mudah Mengoptimumkan Mengoptimumkan itu, membolehkan anda merancang strategi permainan dengan lebih baik, bonus hiburan juga sering kali datang dalam bentuk kredit tambahan atau putaran percuma yang boleh meningkatkan peluang untuk menang, mengapa bonus hiburan menjadi pilihan utama pemain.

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