framework desktop offline ai updates

active/device_framework_desktop/quadlets/ai-external.pod

@@ -0,0 +1,6 @@
+[Pod]
+# ai-external is the primary network
+Network=ai-external.network
+Network=ai-internal.network
+# open-webui
+PublishPort=8080:8080/tcp

active/device_framework_desktop/quadlets/ai-internal.network

@@ -0,0 +1,3 @@
+[Network]
+IPv6=true
+Internal=true

active/device_framework_desktop/quadlets/ai-internal.pod

@@ -0,0 +1,8 @@
+[Pod]
+Network=ai-internal.network
+# llama.cpp
+PublishPort=8000:8000/tcp
+# stable-diffusion.cpp gen
+PublishPort=1234:1234/tcp
+# stable-diffusion.cpp edit
+PublishPort=1235:1235/tcp

active/device_framework_desktop/quadlets/ai.pod

@@ -1,8 +0,0 @@
-[Pod]
-Network=ai.network
-# llama.cpp
-PublishPort=8000:8000/tcp
-# open-webui
-PublishPort=8080:8080/tcp
-# stable-diffusion.cpp
-PublishPort=1234:1234/tcp

active/device_framework_desktop/quadlets/llama-server.container

@@ -2,8 +2,8 @@
 Description=A Llama CPP Server Running GPT OSS 120b
 
 [Container]
-# Shared AI pod
-Pod=ai.pod
+# Shared AI internal pod
+Pod=ai-internal.pod
 
 # Image is built locally via podman build
 Image=localhost/llama-cpp-vulkan:latest
@@ -18,8 +18,6 @@ AddDevice=/dev/dri
 # Server command
 Exec=--port 8000 \
     -c 48000 \
-    -b 48000 \
-    -ub 500 \
     --perf \
     --n-gpu-layers all \
     --jinja \
@@ -27,7 +25,7 @@ Exec=--port 8000 \
     --models-dir /models
 
 # Health Check
-HealthCmd=CMD-SHELL curl --fail http://127.0.0.1:8000/props?model=gpt-oss-120b || exit 1
+HealthCmd=CMD-SHELL curl --fail http://127.0.0.1:8000/props || exit 1
 HealthInterval=10s
 HealthRetries=3
 HealthStartPeriod=10s

active/device_framework_desktop/quadlets/open-webui.container

@@ -2,8 +2,8 @@
 Description=An Open Webui Frontend for Local AI Services
 
 [Container]
-# Shared AI pod
-Pod=ai.pod
+# Shared AI external pod
+Pod=ai-external.pod
 
 # Open Webui base image
 Image=ghcr.io/open-webui/open-webui:main

active/device_framework_desktop/quadlets/stable-diffusion-edit-server.container

@@ -2,8 +2,8 @@
 Description=A Stable Diffusion CPP Server for Editing Images
 
 [Container]
-# Shared AI pod
-Pod=ai.pod
+# Shared AI Internal pod
+Pod=ai-internal.pod
 
 # Vulkan image for AMD GPU
 Image=localhost/stable-diffusion-cpp:latest
@@ -21,16 +21,14 @@ Entrypoint=/sd-server
 # Server args
 Exec=-l 0.0.0.0 \
     --listen-port 1235 \
-    --diffusion-model /models/image/flux-1-kontext/flux1-kontext-dev-Q4_K_M.gguf \
-    --vae /models/image/flux-1-kontext/ae.safetensors \
-    --clip_l /models/image/flux-1-kontext/clip_l.safetensors \
-    --t5xxl /models/image/flux-1-kontext/t5xxl_fp16.safetensors \
+    --diffusion-model /models/image/flux2-klein/flux-2-klein-9b-Q4_0.gguf \
+    --vae /models/image/flux2-klein/ae.safetensors \
+    --llm /models/image/flux2-klein/Qwen3-8B-Q4_K_M.gguf \
     --cfg-scale 1.0 \
     --sampling-method euler \
-    --vae-conv-direct \
-    --seed -1 \
-    --steps 28 \
-    -v
+    -v \
+    --steps 4 \
+    --vae-conv-direct
 
 [Service]
 Restart=always

active/device_framework_desktop/quadlets/stable-diffusion-gen-server.container

@@ -2,8 +2,8 @@
 Description=A Stable Diffusion CPP Server for Generating Images
 
 [Container]
-# Shared AI pod
-Pod=ai.pod
+# Shared AI internal pod
+Pod=ai-internal.pod
 
 # Vulkan image for AMD GPU
 Image=localhost/stable-diffusion-cpp:latest
@@ -24,8 +24,6 @@ Exec=-l 0.0.0.0 \
     --diffusion-model /models/image/z-turbo/z_image_turbo-Q4_K.gguf \
     --vae /models/image/z-turbo/ae.safetensors  \
     --llm /models/image/z-turbo/qwen_3_4b.safetensors \
-    -l 0.0.0.0 \
-    --listen-port 1234 \
     --cfg-scale 1.0 \
     --vae-conv-direct \
     -v \