import torch
from diffusers import DiffusionPipeline
# switch to "mps" for apple devices
pipe = DiffusionPipeline.from_pretrained("kmax002/ideogram-4-fp8-merged", torch_dtype=torch.bfloat16, device_map="cuda")
prompt = "Astronaut in a jungle, cold color palette, muted colors, detailed, 8k"
image = pipe(prompt).images[0]YAML Metadata Warning:empty or missing yaml metadata in repo card
Check out the documentation for more information.
ideogram-4-fp8-merged
NOTE: weights are now bf16, not fp8. Ideogram-4 with the twinks_10000 LoKr merged in at full bf16 precision (crisp, no fp8 requantization). Layout matches ideogram-ai/ideogram-4-fp8 (transformer/ + unconditional_transformer/), but each diffusion_pytorch_model.safetensors is bf16, single-shard, 458 tensors, no weight_scale. Text encoder / VAE / tokenizer / scheduler are the stock Ideogram-4 files. For AI Toolkit, set the Ideogram model name_or_path to this repo.
- Downloads last month
- 3