fix: use q5_0 quantization type (q5_k_m not supported by sd.cpp)

Dockerfile

@@ -44,8 +44,8 @@ RUN curl -fL --retry 3 --retry-delay 5 -o /app/models/z-anime-distill-4step-bf16
     "https://huggingface.co/SeeSee21/Z-Anime/resolve/main/diffusion_models/z-anime-distill-4step-bf16.safetensors" \
     && /app/sd-cli -M convert \
         -m /app/models/z-anime-distill-4step-bf16.safetensors \
-        -o /app/models/z-anime-distill-4step-Q5_K_M.gguf \
-        --type q5_k_m \
+        -o /app/models/z-anime-distill-4step-Q5_0.gguf \
+        --type q5_0 \
     && rm /app/models/z-anime-distill-4step-bf16.safetensors
 
 # Download Qwen3-4B text encoder Q8_0 GGUF (~4.28GB)

app.py

@@ -7,7 +7,7 @@ import gradio as gr
 # ---------------------------------------------------------------------------
 # Model paths (downloaded at build time)
 # ---------------------------------------------------------------------------
-DIFFUSION = "/app/models/z-anime-distill-4step-Q5_K_M.gguf"
+DIFFUSION = "/app/models/z-anime-distill-4step-Q5_0.gguf"
 LLM = "/app/models/qwen3_4b_q8_0.gguf"
 VAE = "/app/models/ae.safetensors"