Update README.md

README.md

@@ -1,56 +1,39 @@
 ---
+license: openrail++
 library_name: diffusers
 pipeline_tag: text-to-image
 tags:
+  - stable-diffusion
   - stable-diffusion-xl
+  - sdxl
   - diffusers
   - text-to-image
   - image-generation
-  - sdxl
   - lora
-  - merged
-  - pipeline
-
-model-index:
-  - name: Mythic SDXL (LoRA Merged)
-    results:
-      - task:
-          type: text-to-image
-          name: Text-to-Image Generation
-        dataset:
-          name: internal-eval
-          type: subjective-evaluation
-        metrics:
-          - type: visual-quality
-            value: "Passed visual inspection"
-            name: Visual Quality (Subjective)
+inference: true
+base_model: stabilityai/stable-diffusion-xl-base-1.0
 ---
 
-# 🧙 Mythic SDXL — LoRA Merged (Diffusers)
+# Mythic SDXL (LoRA Merged) — Inference Ready
 
-**Mythic SDXL** is a fully merged Stable Diffusion XL pipeline, combining:
+This model is a **Stable Diffusion XL 1.0 pipeline merged with a custom LoRA**, exported in full Diffusers format and ready for Hugging Face **Inference API**.
 
-| Component | Source |
-|----------|--------|
-| Base Model | `stabilityai/stable-diffusion-xl-base-1.0`
-| LoRA | `CoolKrishh/Comic-SDXL-LoRA`
-| Merge Style | `pipe.load_lora_weights()` + `pipe.fuse_lora()`
+## ✅ Supports
+- Hugging Face **Text-to-Image Inference API**
+- `InferenceClient.text_to_image`
+- Diffusers `StableDiffusionXLPipeline`
 
-✅ No LoRA needed during inference  
-✅ No `.safetensors` loading  
-✅ Works directly through **Hugging Face Inference API**
+## 🧬 Model details
+- **Base model:** SDXL 1.0
+- **Finetune method:** LoRA (merged into UNet + Text Encoders)
+- **Format:** Full Diffusers pipeline (not standalone safetensors)
+- **Precision:** FP16
 
----
-
-## 🚀 Quick Start — Hugging Face Inference API
+## 🚀 Quick Inference
 
 ```python
 from huggingface_hub import InferenceClient
 
 client = InferenceClient("CoolKrishh/mythic-sdxl")
-
-image = client.text_to_image(
-    "mythic artwork, cinematic lighting, ultra detailed, 8k"
-)
-
-image.save("output.png")
+img = client.text_to_image("dark comic hero with red lightning")
+img.save("output.png")