update README

README.md

@@ -1,13 +1,50 @@
-# Container Repository for CPU adaptations of Inference code
+# Model Repository for Big Endian models
 
 ## Variants
 
+### Arco 500M Q4 BE
+
+
+# Container Repository for CPU adaptations of Inference code
+
+## Variants for Inference
+
+### Slim container
+
+- run std binaries
+
+
 ### CPUdiffusion
 
 - inference diffusion models on CPU
 - include CUDAonCPU stack
 
-### CPUgguf
+### Diffusion container
+
+- run diffusion app.py variants
+- support CPU and CUDA
+- include Flux
+
+### Slim CUDA container
+
+- run CUDA binaries
+
+
+## Variants for Build
+
+### Llama.cpp build container
+
+- build llama-cli-static
+- build llama-server-static
+
+### sd build container
+
+- build sd
+- optional: build sd-server
+
+### CUDA build container
+
+- build cuda binaries
+- support sd_cuda
+
 
-- inference gguf models on CPU
-- include GUI libraries

serve-from-url.sh

@@ -0,0 +1,10 @@
+		1) wget the model
+		2) save hostip:		ip route | grep 'default' | awk '{print $9}' >hostip
+		3a) calls llama-server in container
+		3b) calls sed + llama-server in container
+	
+		opodman run -d --net=host -v ~/funstreams:/models localhost/bookworm:server ./models/llama-server-static -m /models/qwen2-500.gguf
+
+
+
+