Z-Image ยท FP8 (transformer)

FP8 (float8_e4m3fn) quantization of the Z-Image DiT (ZImageTransformer2DModel, 6.2B) transformer from Tongyi-MAI/Z-Image. Quantized by axteam.

File Size
Z-Image-transformer-fp8_e4m3fn.safetensors 6.2 GB

What this is

  • Only the transformer (DiT) is quantized to float8_e4m3fn (2-D weight tensors; norms/biases kept in BF16). ~half the size of the BF16 transformer.
  • VAE and text encoder(s) are NOT included โ€” use the ones from the original Tongyi-MAI/Z-Image.

Checked on GPU (diffusers)

On 2026-11-02 the FP8 values of this file (the same values ComfyUI loads; upcast to BF16 for this run) were loaded into the transformer of the original Tongyi-MAI/Z-Image pipeline (diffusers 0.42.1, NVIDIA B200; all tensor names matched the original transformer โ€” 0 missing, 0 unexpected) and compared with the original pipeline run in BF16 at the same prompt and seed (1024ร—1024, the pipeline's default sampler settings).

Prompt Pixel PSNR vs original (BF16 run) Mean abs. pixel diff
A red apple on a wooden table, studio lighting, photorealistic 25.5 dB 9.4 / 255
A cozy cafe street in Seoul at night with warm lights, illustration 14.7 dB 30.9 / 255

Both images show the same subject but are less clean than the BF16 ones: the apple keeps a similar composition (shorter stem, slightly different shape), with faint blocky patches in the blurred background and light smudges on the table that the BF16 image does not have; the detailed street scene changes its layout, and its lit pavement shows faint streaks. Pixel PSNR is strict for diffusion models: a small weight change shifts the sampling path, so a detailed scene can change its layout while the subject stays correct.

Requirements

Runtime ComfyUI (Load Diffusion Model node, float8_e4m3fn single file). Not run in ComfyUI here (no ComfyUI on this server); the FP8 weights were checked with diffusers โ€” see Checked on GPU above.
Also needed the Qwen3 text encoder (8.0 GB) and VAE (0.17 GB) from the base repo.
Memory this file is 6.2 GB; add the text encoder(s) and VAE above.

Usage (ComfyUI)

  1. Download the FP8 file into ComfyUI's diffusion-model folder:
hf download axteam/Z-Image-FP8 Z-Image-transformer-fp8_e4m3fn.safetensors --local-dir ComfyUI/models/diffusion_models
  1. Get the VAE and text encoder(s) from the original Tongyi-MAI/Z-Image repo (or the matching ComfyUI repackaged files) and put them in ComfyUI/models/vae/ and ComfyUI/models/text_encoders/.
  2. In your workflow, load this file with the Load Diffusion Model node (UNETLoader) โ€” leave weight_dtype at default (the file is already float8_e4m3fn) โ€” and wire the original VAE / text-encoder loaders as in the base model's reference workflow.
  3. Sampler settings (steps, CFG/guidance, resolution) are the same as for the original model.

The FP8 file is a plain cast of the transformer weights (no scaling), the common ComfyUI fp8_e4m3fn convention; norms and biases stay BF16.

License

Quantized derivative of Tongyi-MAI/Z-Image. Original weights belong to the upstream authors; the base model's license applies to this derivative.


ํ•œ๊ตญ์–ด ์š”์•ฝ

Tongyi-MAI/Z-Image์˜ transformer(DiT) FP8 ์–‘์žํ™” ๋นŒ๋“œ (ComfyUI์šฉ ๋‹จ์ผ ํŒŒ์ผ) ยท ์ œ์ž‘ axteam

๊ฐœ์š”

ํ•ญ๋ชฉ ๋‚ด์šฉ
์›๋ณธ ๋ชจ๋ธ Tongyi-MAI/Z-Image
ํ˜•์‹ float8_e4m3fn ๋‹จ์ผ safetensors (ComfyUI์—์„œ ํ”ํžˆ ์“ฐ๋Š” FP8 ํ˜•์‹)
ํŒŒ์ผ Z-Image-transformer-fp8_e4m3fn.safetensors 6.2 GB
ํฌํ•จ transformer(DiT)

์–‘์žํ™” ์ƒ์„ธ

  • ์–‘์žํ™” ์ ˆ์ฐจ: transformer์˜ 2D ๊ฐ€์ค‘์น˜๋ฅผ ์Šค์ผ€์ผ ์—†์ด float8_e4m3fn์œผ๋กœ ๋‹จ์ˆœ ํ˜•๋ณ€ํ™˜(cast) โ€” ComfyUI์—์„œ ํ”ํžˆ ์“ฐ๋Š” ๋ฐฉ์‹
  • BF16 ์œ ์ง€: ์ •๊ทœํ™”์ธต, bias
  • ํฌ๊ธฐ: BF16 transformer์˜ ์•ฝ ์ ˆ๋ฐ˜
  • ๋ฏธํฌํ•จ: VAEยทํ…์ŠคํŠธ ์ธ์ฝ”๋” โ€” ์›๋ณธ ๋ฆฌํฌ์—์„œ ๋ฐ›์•„ ์‚ฌ์šฉ
  • ํ’ˆ์งˆ(์‹ค์ธก, 2026-11-02): diffusers 0.42.1๋กœ ์›๋ณธ ํŒŒ์ดํ”„๋ผ์ธ์„ BF16์œผ๋กœ ์‹คํ–‰ํ•œ ๊ทธ๋ฆผ๊ณผ, ๊ฐ™์€ ํŒŒ์ดํ”„๋ผ์ธ์˜ transformer์— ์ด ํŒŒ์ผ์˜ FP8 ๊ฐ’(ComfyUI๊ฐ€ ์ฝ๋Š” ๊ฒƒ๊ณผ ๊ฐ™์€ ๊ฐ’, ์ด ์ ๊ฒ€์—์„œ๋Š” BF16์œผ๋กœ ์˜ฌ๋ ค ๊ณ„์‚ฐ)์„ ๋„ฃ์–ด ๊ทธ๋ฆฐ ๊ทธ๋ฆผ์„ ๊ฐ™์€ ํ”„๋กฌํ”„ํŠธยท์‹œ๋“œ 2๊ฐœ๋กœ ๋น„๊ต(1024ร—1024) โ€” ๋‘ ๊ทธ๋ฆผ ๋ชจ๋‘ ๊ฐ™์€ ์ฃผ์ œ์ด๋‚˜ ์›๋ณธ๋ณด๋‹ค ๋œ ๊นจ๋—ํ•จ, ์‚ฌ๊ณผ๋Š” ๋น„์Šทํ•œ ๊ตฌ๋„(๊ผญ์ง€๊ฐ€ ์งง๊ณ  ๋ชจ์–‘์ด ์กฐ๊ธˆ ๋‹ค๋ฆ„)์ด๋‚˜ ํ๋ฆฐ ๋ฐฐ๊ฒฝ์— ์›๋ณธ์— ์—†๋Š” ์˜…์€ ๋„ค๋ชจ ์–ผ๋ฃฉ๊ณผ ํƒ์ž ์œ„ ๋ฐ์€ ์ž๊ตญ์ด ๋ณด์ด๊ณ , ๋ณต์žกํ•œ ๊ฑฐ๋ฆฌ ์žฅ๋ฉด์€ ๊ตฌ๋„๊ฐ€ ๋‹ฌ๋ผ์ง€๋ฉฐ ๋ถˆ๋น› ๋น„์นœ ๋ฐ”๋‹ฅ์— ์˜…์€ ์ค„๋ฌด๋Šฌ๊ฐ€ ๋ณด์ž„, ํ”ฝ์…€ PSNR 25.5ยท14.7 dB (๋””ํ“จ์ „์€ ๊ฐ€์ค‘์น˜๊ฐ€ ์กฐ๊ธˆ๋งŒ ๋‹ฌ๋ผ๋„ ์ƒ์„ฑ ๊ฒฝ๋กœ๊ฐ€ ๋ฐ”๋€Œ์–ด ๋ณต์žกํ•œ ์žฅ๋ฉด์€ ๊ตฌ๋„๊ฐ€ ๋‹ฌ๋ผ์งˆ ์ˆ˜ ์žˆ์Œ, ์œ„ ์˜์–ด ๋ณธ๋ฌธ ํ‘œ ์ฐธ๊ณ )

์š”๊ตฌ์‚ฌํ•ญ

  • ์‹คํ–‰ ํ™˜๊ฒฝ: ComfyUI โ€” Load Diffusion Model ๋…ธ๋“œ๋กœ float8_e4m3fn ๋‹จ์ผ ํŒŒ์ผ ๋กœ๋“œ (ComfyUI๋กœ๋Š” ์‹คํ–‰ ๋ฏธํ™•์ธ โ€” FP8 ๊ฐ€์ค‘์น˜๋Š” diffusers๋กœ ์ƒ์„ฑ ํ™•์ธ, ํ’ˆ์งˆ ํ•ญ๋ชฉ ์ฐธ๊ณ )
  • ์ถ”๊ฐ€๋กœ ํ•„์š”ํ•œ ํŒŒ์ผ: ์›๋ณธ ๋ฆฌํฌ์˜ Qwen3 ํ…์ŠคํŠธ ์ธ์ฝ”๋”(8.0 GB)์™€ VAE(0.17 GB)
  • ๋ฉ”๋ชจ๋ฆฌ: ์ด ํŒŒ์ผ 6.2 GB + ์œ„ ํ…์ŠคํŠธ ์ธ์ฝ”๋”ยทVAE

์‚ฌ์šฉ๋ฒ• (ComfyUI)

  • ํŒŒ์ผ ๋ฐ›๊ธฐ: hf download axteam/Z-Image-FP8 Z-Image-transformer-fp8_e4m3fn.safetensors --local-dir ComfyUI/models/diffusion_models
  • ํ…์ŠคํŠธ ์ธ์ฝ”๋”ยทVAE: ์›๋ณธ ๋ฆฌํฌ์—์„œ ๋ฐ›์•„ ComfyUI/models/text_encoders/ยทComfyUI/models/vae/์— ๋ฐฐ์น˜
  • ๋ถˆ๋Ÿฌ์˜ค๊ธฐ: Load Diffusion Model ๋…ธ๋“œ(UNETLoader), weight_dtype์€ default (ํŒŒ์ผ์ด ์ด๋ฏธ float8_e4m3fn)
  • ์ƒ˜ํ”Œ๋Ÿฌ ์„ค์ •: ์Šคํ…ยทCFG(guidance)ยทํ•ด์ƒ๋„๋Š” ์›๋ณธ ๋ชจ๋ธ๊ณผ ๋™์ผ

๋ผ์ด์„ ์Šค

  • ๋ผ์ด์„ ์Šค: Apache-2.0 โ€” ์›๋ณธ ๋ชจ๋ธ์˜ ๋ผ์ด์„ ์Šค๋ฅผ ๊ทธ๋Œ€๋กœ ๋”ฐ๋ฆ„
  • ๊ถŒ๋ฆฌ: ์›๋ณธ ๊ฐ€์ค‘์น˜๋Š” Tongyi-MAI(Alibaba) ์†Œ์œ , ์ด ๋ฆฌํฌ๋Š” ์–‘์žํ™” ํŒŒ์ƒ๋ฌผ
Downloads last month

-

Downloads are not tracked for this model. How to track
Inference Providers NEW
This model isn't deployed by any Inference Provider. ๐Ÿ™‹ Ask for provider support

Model tree for axteam/Z-Image-FP8

Quantized
(28)
this model

Collection including axteam/Z-Image-FP8