Instructions to use SUSSYEATING/GLM-Image-ConvRot with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- Diffusers
How to use SUSSYEATING/GLM-Image-ConvRot with Diffusers:
pip install -U diffusers transformers accelerate
import torch from diffusers import DiffusionPipeline # switch to "mps" for apple devices pipe = DiffusionPipeline.from_pretrained("SUSSYEATING/GLM-Image-ConvRot", dtype=torch.bfloat16, device_map="cuda") prompt = "Astronaut in a jungle, cold color palette, muted colors, detailed, 8k" image = pipe(prompt).images[0] - Notebooks
- Google Colab
- Kaggle
- Local Apps Settings
- Draw Things
- DiffusionBee
|
Download README.md from SUSSYEATING/GLM-Image-ConvRot: direct link, hf CLI and curl.
- Browser
- Download file 5.06 kB
-
https://huggingface.co/SUSSYEATING/GLM-Image-ConvRot/resolve/main/README.md
- Command line
-
hf download hf://SUSSYEATING/GLM-Image-ConvRot/README.md
-
curl -L -o README.md https://huggingface.co/SUSSYEATING/GLM-Image-ConvRot/resolve/main/README.md
5.06 kB
metadata
license: mit
language:
- zh
- en
library_name: diffusers
pipeline_tag: text-to-image
base_model: zai-org/GLM-Image
tags:
- glm-image
- convrot
- quantization
- int8
- int6
- int4
- comfy-kitchen
GLM-Image ConvRot (INT8, INT6 w6w8, INT4 w4w8)
High-fidelity ConvRot quantized checkpoints for zai-org/GLM-Image (hybrid 9B Autoregressive VLM + 7B Diffusion Decoder DiT), powered by comfy-kitchen INT8 / W6A8 / W4A8 Tensor Core kernels.
- 7B DiT (
transformer): All180attention/FFN block linears quantized with ConvRot-256 (int8_tensorwise,w6a8_int8,asym_w4a8_int8) +30AdaLN modulation layers (norm1.linear) kept atINT8 ConvRot-256across all tiers;prior_token_embedding,glyph_projector,image_projector,time_condition_embed,norm_out, andproj_outremain BF16. - 9B AR VLM (
vision_language_encoder): All40AR LLM layers quantized with SmoothQuant + ConvRot-256 (q_proj,o_proj,mlp.gate_up_proj) andINT8 ConvRot-64onmlp.down_proj(K = 13696, divisible by 64) forw6w8/w4w8compatibility;40ViT blocks quantized toINT8 ConvRot-256;k_proj,v_proj(0.17 GBtotal),embed_tokens,lm_head, andvqmodelremain BF16 to preserve KV-cache precision and discrete visual-token accuracy. - Typography & VAE (
text_encoderByT5 +vae): Loaded unquantized directly fromzai-org/GLM-Image(1.68 GBcombined) for crisp text rendering and pixel fidelity.
Verified Benchmarks (1024x1024, 50 DiT steps, guidance_scale=1.5)
| Tier | DiT + VLM Size | Peak VRAM | DiT Warm Time | DiT-Only PSNR (cabin / poster / watch) |
DiT-Only SSIM | VLM Teacher-Forced Top-1 / KL |
|---|---|---|---|---|---|---|
| BF16 Baseline | 34.08 GB (13.85 + 20.23) |
40.55 GB |
21.60s |
Ref | 1.0000 |
100.0% / 0.000 |
| INT8 ConvRot | 18.10 GB (7.07 + 11.03, -46.9%) |
24.59 GB |
17.21s (1.26x) |
47.21 / 42.83 / 38.65 dB |
0.9982 / 0.9965 / 0.9957 |
89.71% / 0.081 |
INT6 (w6w8) ConvRot |
15.88 GB (5.94 + 9.94, -53.4%) |
22.31 GB |
17.95s (1.20x) |
41.74 / 39.73 / 36.68 dB |
0.9966 / 0.9941 / 0.9933 |
89.19% / 0.082 |
INT4 (w4w8) ConvRot |
12.91 GB (4.43 + 8.48, -62.1%) |
19.40 GB |
17.80s (1.21x) |
30.44 / 30.55 / 26.37 dB |
0.9723 / 0.9650 / 0.9608 |
79.62% / 0.316 |
Note: In end-to-end sampling, GLM-Image's AR VLM autoregressively samples 4,096 discrete visual tokens, so small logit shifts produce different valid compositions of the same prompt. Both DiT-isolated reconstruction (identical BF16 prior tokens -> quantized DiT) and VLM teacher-forced visual-token agreement are reported above alongside full end-to-end visual comparisons below.
Visual Comparisons (BF16 | INT8 | INT6 w6w8 | INT4 w4w8)
End-to-End Photorealistic Scene (t2i_cabin_s42)
End-to-End Typography Poster (t2i_poster_s1337)
End-to-End Macro Detail (t2i_watch_s777)
DiT-Isolated Reconstruction Fidelity (t2i_cabin_s42, Same BF16 Prior Tokens)
Quickstart (diffusers + comfy-kitchen)
pip install comfy-kitchen>=0.2.41 diffusers transformers safetensors accelerate
import torch
from huggingface_hub import hf_hub_download, snapshot_download
from glm_image_convrot import load_glm_image_convrot
base_model = snapshot_download("zai-org/GLM-Image")
tier = "int6" # "int8", "int6", or "int4"
transformer_path = hf_hub_download(
"SUSSYEATING/GLM-Image-ConvRot",
f"{tier}/glm-image-transformer-{tier}-convrot.safetensors",
)
vlm_path = hf_hub_download(
"SUSSYEATING/GLM-Image-ConvRot",
f"{tier}/glm-image-vlm-{tier}-convrot.safetensors",
)
pipe, info = load_glm_image_convrot(
base_model=base_model,
transformer_path=transformer_path,
vlm_path=vlm_path,
device="cuda",
dtype=torch.bfloat16,
)
image = pipe(
prompt="A vintage travel poster for 'ALPS EXPRESS' with bold cream typography, a red steam train crossing a stone viaduct over snowy mountains, 'SWITZERLAND 1964'",
height=1024,
width=1024,
num_inference_steps=50,
guidance_scale=1.5,
generator=torch.Generator(device="cuda").manual_seed(1337),
).images[0]
image.save("glm_convrot_out.png")
License & Attribution
- Derived from zai-org/GLM-Image, released under the MIT License.
- As noted upstream, the incorporated VQ tokenizer and ViT weights from X-Omni/X-Omni-En remain subject to the Apache License, Version 2.0.



