new

.gitattributes

@@ -1,35 +0,0 @@
-*.7z filter=lfs diff=lfs merge=lfs -text
-*.arrow filter=lfs diff=lfs merge=lfs -text
-*.bin filter=lfs diff=lfs merge=lfs -text
-*.bz2 filter=lfs diff=lfs merge=lfs -text
-*.ckpt filter=lfs diff=lfs merge=lfs -text
-*.ftz filter=lfs diff=lfs merge=lfs -text
-*.gz filter=lfs diff=lfs merge=lfs -text
-*.h5 filter=lfs diff=lfs merge=lfs -text
-*.joblib filter=lfs diff=lfs merge=lfs -text
-*.lfs.* filter=lfs diff=lfs merge=lfs -text
-*.mlmodel filter=lfs diff=lfs merge=lfs -text
-*.model filter=lfs diff=lfs merge=lfs -text
-*.msgpack filter=lfs diff=lfs merge=lfs -text
-*.npy filter=lfs diff=lfs merge=lfs -text
-*.npz filter=lfs diff=lfs merge=lfs -text
-*.onnx filter=lfs diff=lfs merge=lfs -text
-*.ot filter=lfs diff=lfs merge=lfs -text
-*.parquet filter=lfs diff=lfs merge=lfs -text
-*.pb filter=lfs diff=lfs merge=lfs -text
-*.pickle filter=lfs diff=lfs merge=lfs -text
-*.pkl filter=lfs diff=lfs merge=lfs -text
-*.pt filter=lfs diff=lfs merge=lfs -text
-*.pth filter=lfs diff=lfs merge=lfs -text
-*.rar filter=lfs diff=lfs merge=lfs -text
-*.safetensors filter=lfs diff=lfs merge=lfs -text
-saved_model/**/* filter=lfs diff=lfs merge=lfs -text
-*.tar.* filter=lfs diff=lfs merge=lfs -text
-*.tar filter=lfs diff=lfs merge=lfs -text
-*.tflite filter=lfs diff=lfs merge=lfs -text
-*.tgz filter=lfs diff=lfs merge=lfs -text
-*.wasm filter=lfs diff=lfs merge=lfs -text
-*.xz filter=lfs diff=lfs merge=lfs -text
-*.zip filter=lfs diff=lfs merge=lfs -text
-*.zst filter=lfs diff=lfs merge=lfs -text
-*tfevents* filter=lfs diff=lfs merge=lfs -text

README.md

@@ -1,18 +1 @@
----
-title: Minecraftify
-emoji: ⚡
-colorFrom: blue
-colorTo: red
-sdk: gradio
-sdk_version: 6.18.0
-python_version: '3.13'
-app_file: app.py
-pinned: false
-license: apache-2.0
-short_description: Mincraftify converts all images into mc-style LIVE!
----
-
-# ⛏️ Minecraft Spatial Voxel Filter
-Real-time video stream processing pipeline using FastRTC and Modal serverless environments.
-
-Check out the configuration reference at https://huggingface.co/docs/hub/spaces-config-reference
+# Minecrafity

backend/backend.py

@@ -1,127 +0,0 @@
-import io
-import os
-import modal
-
-# ==============================================================================
-# 💾 0. CACHE VOLUME SETUP
-# ==============================================================================
-# This creates a persistent volume that survives between container cold-starts.
-cache_volume = modal.Volume.from_name("flux-inductor-cache", create_if_missing=True)
-CACHE_DIR = "/root/.cache/torch/inductor"
-
-# Define container environment
-image = (
-    modal.Image.debian_slim(python_version="3.12")
-    .pip_install(
-        "diffusers",
-        "transformers",
-        "accelerate",
-        "pillow",
-        "torch",
-        "triton"  # Essential for torch.compile
-    )
-    # Tell PyTorch Inductor to write its cache to our persistent volume directory
-    .env({"TORCHINDUCTOR_CACHE_DIR": CACHE_DIR})
-)
-
-app = modal.App("flux-klein-voxel-backend", image=image)
-
-# ==============================================================================
-# 🏎️ 1. THE DEMO PIPELINE (FALLBACK ROUTE)
-# ==============================================================================
-@app.function()
-def demo_stream_frame(img_bytes: bytes) -> bytes:
-    """Fallback route structurally aligned to match the frontend signature."""
-    from PIL import Image, ImageDraw
-    
-    input_image = Image.open(io.BytesIO(img_bytes)).convert("RGB")
-    draw = ImageDraw.Draw(input_image)
-    draw.text((20, 20), "🛠️ WEBRTC PASSTHROUGH DEMO ACTIVE", fill=(0, 255, 0))
-    
-    output_buffer = io.BytesIO()
-    input_image.save(output_buffer, format="JPEG", quality=85)
-    return output_buffer.getvalue()
-
-
-# ==============================================================================
-# 🚀 2. THE REAL-TIME VOXEL ENGINE
-# ==============================================================================
-@app.cls(
-    gpu="A10G",
-    secrets=[modal.Secret.from_name("huggingface-secret")],
-    max_containers=5,
-    # Mount the persistent cache volume to the exact path PyTorch is looking at
-    volumes={CACHE_DIR: cache_volume} 
-)
-class VoxelModel:
-    
-    @modal.enter()
-    def load_pipeline(self):
-        import torch
-        from PIL import Image
-        from diffusers import AutoPipelineForImage2Image
-        
-        model_id = "AnimeOverlord/flux2-klein-4b-mc"
-        
-        self.pipe = AutoPipelineForImage2Image.from_pretrained(
-            model_id,
-            torch_dtype=torch.bfloat16,
-            token=os.environ["HF_TOKEN"]  # Modernized from use_auth_token
-        )
-        self.pipe.to("cuda")
-        self.pipe.enable_attention_slicing()
-
-        # ---------------------------------------------------------
-        # ✨ COMPILE AND OPTIMIZE THE INFRASTRUCTURE
-        # ---------------------------------------------------------
-        print("Initializing torch.compile optimization loops...")
-        
-        # Channels-last optimization applied exclusively to the VAE (CNN-based)
-        self.pipe.vae.to(memory_format=torch.channels_last)
-        
-        # Compile the heaviest part of the FLUX architecture safely without memory layout issues
-        self.pipe.transformer = torch.compile(
-            self.pipe.transformer,
-            mode="reduce-overhead",  # Trades a bit of compile time for faster inference
-            fullgraph=False
-        )
-
-        # WARMUP RUN: Force a dummy inference immediately execution starts.
-        # If the cache is empty (first run), this traces the graph and saves to the Modal Volume.
-        # If the cache exists, it instantly hotloads from the Volume.
-        print("Running warmup to build/load inductor cache...")
-        dummy_image = Image.new("RGB", (512, 512), (0, 0, 0))
-        with torch.inference_mode():
-            self.pipe(
-                prompt="warmup pass",
-                image=dummy_image,
-                strength=0.5,
-                num_inference_steps=2,
-                guidance_scale=1.0,
-            )
-        print("Warmup complete. Ready for real-time requests!")
-
-    @modal.method()
-    def process_frame(self, img_bytes: bytes) -> bytes:
-        from PIL import Image
-        import torch
-        
-        # Hardcoded parameters that were previously passed from the UI
-        prompt = "isometric 3d minecraft block voxel style, high resolution, volumetric lighting"
-        strength = 0.45
-        
-        input_image = Image.open(io.BytesIO(img_bytes)).convert("RGB")
-        input_image = input_image.resize((512, 512)) 
-        
-        with torch.inference_mode():
-            output_image = self.pipe(
-                prompt=prompt,
-                image=input_image,
-                strength=strength,
-                num_inference_steps=4,
-                guidance_scale=3.5,
-            ).images[0]
-            
-        output_buffer = io.BytesIO()
-        output_image.save(output_buffer, format="JPEG", quality=85)
-        return output_buffer.getvalue()