Added GLM and Qwen3.5cpu

GLM-4.7-Flash-Uncensored.sh

@@ -0,0 +1,34 @@
+#!/bin/bash
+#grab container image name
+MODEL_NAME="GLM-4.7-Flash-Uncen-Hrt"
+CONT_NAME="LLM-$MODEL_NAME"
+
+MODEL_FOLDER="/home/jared/.cache"
+MODEL="huggingface/hub/models--DavidAU--GLM-4.7-Flash-Uncensored-Heretic-NEO-CODE-Imatrix-MAX-GGUF/snapshots/af1cb75ecde13d21c19ceaa5fbd4352f8233e0e5/GLM-4.7-Flash-Uncen-Hrt-NEO-CODE-MAX-imat-D_AU-Q8_0.gguf"
+#MODEL_MM="huggingface/hub/models--HauhauCS--Qwen3.6-35B-A3B-Uncensored-HauhauCS-Aggressive/snapshots/f12a584fecbeb5f20001130d8ecd66c9327ae685/mmproj-Qwen3.6-35B-A3B-Uncensored-HauhauCS-Aggressive-f16.gguf"
+
+PORT=8356
+
+# Check if llama_env.sh exists, if not run create_new_image.sh
+if [ ! -f "./env/llama_env.sh" ]; then
+    echo "llama_env.sh not found, running create_new_image.sh..."
+    ./create_new_image.sh
+fi
+source ./env/llama_env.sh
+echo $IMAGE
+docker container create --name $CONT_NAME --network llms -p $PORT:$PORT -v $MODEL_FOLDER:/model --user $(id -u):$(id -g) --gpus=all --restart on-failure:3 \
+	--health-cmd "curl -f http://$CONT_NAME:$PORT/v1/models || exit 1" \
+	--health-interval 5s \
+	--health-timeout 5s \
+	--health-start-period 20s \
+	$IMAGE \
+	/llama.cpp/build/bin/llama-server \
+	--model /model/$MODEL \
+	--alias $MODEL_NAME \
+	-ctk q8_0 -ctv q8_0 \
+	--host $CONT_NAME \
+	--port $PORT --no-mmap -dio
+
+
+echo "Done!"
+echo $CONT_NAME

README.md

@@ -1,8 +1,3 @@
 # ik_llama_docker
 
-Builds docker images to run ik_llama quantized models.
-
-## How to
-First generate the base docker image by running ./create_new_image.sh
-
-Run the specific model script to create the docker container ./qwen3.5-122b.sh
+Builds docker images to run ik_llama quantized models, llama.cpp compatable models, and vllm backend models.

qwen3.5-397B-A17B-IQ2_KL-CPU.sh

@@ -0,0 +1,36 @@
+#!/bin/bash
+#grab container image name
+MODEL_NAME="Qwen3.5-397B-A17B"
+CONT_NAME="LLM-$MODEL_NAME"
+
+MODEL="huggingface/hub/models--ubergarm--Qwen3.5-397B-A17B-GGUF/snapshots/bb9c5e48cdcad8716ef0dfff1f48f43b34b89cb2/IQ2_KL/Qwen3.5-397B-A17B-IQ2_KL-00001-of-00004.gguf"
+MODEL_FOLDER="/home/jared/.cache"
+
+PORT=10987
+
+# Check if llama_env.sh exists, if not run create_new_image.sh
+if [ ! -f "./env/ik_env.sh" ]; then
+    echo "./env/ik_env.sh not found, running create_new_IK_image.sh..."
+    ./create_new_IK_image.sh
+fi
+source ./env/ik_env.sh
+echo $IMAGE
+docker container create --name $CONT_NAME --network llms -p $PORT:$PORT -v $MODEL_FOLDER:/model --user $(id -u):$(id -g) --gpus=all --restart on-failure:3 \
+	--health-cmd "curl -f http://$CONT_NAME:$PORT/v1/models || exit 1" \
+	--health-interval 5s \
+	--health-timeout 5s \
+	--health-start-period 50s \
+	$IMAGE \
+	/ik_llama.cpp/build/bin/llama-server \
+	--model /model/$MODEL \
+	--alias $MODEL_NAME \
+	-ctk q8_0 -ctv q8_0 \
+	-ngl 999 \
+	--parallel 2 --threads 24 \
+	--host $CONT_NAME \
+	--port $PORT --jinja --no-mmap
+
+echo "Done!"
+echo $CONT_NAME
+
+docker container start -a $CONT_NAME