Create README.md

README.md

@@ -0,0 +1,45 @@
+---
+base_model_relation: quantized
+library_name: diffusers
+tags:
+- sdnq
+- ideogram
+- ideogram_4
+- 8-bit
+base_model:
+- ideogram-ai/ideogram-4-fp8
+---
+
+This model is a direct conversion of [Ideogram-4 FP8](https://huggingface.co/ideogram-ai/ideogram-4-fp8) to [SDNQ](https://github.com/Disty0/sdnq) Diffusers format with identical weights from the original FP8 model.  
+ 
+
+```py
+import os
+import json
+import requests
+
+import torch
+import diffusers
+from sdnq import SDNQConfig # import sdnq to register it into diffusers and transformers
+
+pipe = diffusers.Ideogram4Pipeline.from_pretrained("Disty0/Ideogram-4-SDNQ-FP8", torch_dtype=torch.bfloat16)
+pipe.enable_model_cpu_offload()
+
+# Expand the prompt into a structured JSON caption with Ideogram's free hosted magic-prompt API.
+# Get a key at https://developer.ideogram.ai/  (set IDEOGRAM_API_KEY).
+resp = requests.post(
+    "https://api.ideogram.ai/v1/ideogram-v4/magic-prompt",
+    headers={"Api-Key": "your_ideogram_api_key"},
+    json={"text_prompt": "a ginger cat wearing a tiny wizard hat reading a spellbook", "aspect_ratio": "1x1"},
+).json()
+caption = json.dumps(resp["json_prompt"])  # or: token="hf_xxxxxxxxx", token is needed as the repo is gated
+
+# Pass the caption straight to the pipeline (no prompt_upsampling — it's already upsampled).
+image = pipe(
+    caption, 
+    height=1024, # model supports up to 2048
+    width=1024, # model supports up to 2048
+    generator=torch.manual_seed(0),
+).images[0]
+image.save("ideogram4-sdnq-fp8.png")
+```