update model

app.py

@@ -1,5 +1,13 @@
+from huggingface_hub import hf_hub_download, snapshot_download
 from ktem.main import App
 
+# preload model
+snapshot_download("mixedbread-ai/mxbai-embed-large-v1")
+hf_hub_download(
+    repo_id="Qwen/Qwen1.5-7B-Chat-GGUF",
+    filename="qwen1_5-7b-chat-q5_k_m.gguf",
+)
+
 app = App()
 demo = app.make()
 demo.launch()

flowsettings.py

@@ -124,6 +124,18 @@ if len(KH_EMBEDDINGS) < 1:
         "default": True,
     }
 
+
+KH_LLMS["qwen1.5-7B"] = {
+    "spec": {
+        "__type__": "kotaemon.llms.LlamaCppChat",
+        "repo_id": "Qwen/Qwen1.5-7B-Chat-GGUF",
+        "filename": "qwen1_5-7b-chat-q5_k_m.gguf",
+    },
+    "default": False,
+    "cost": 0,
+}
+
+
 KH_REASONINGS = ["ktem.reasoning.simple.FullQAPipeline"]
 KH_VLM_ENDPOINT = "{0}/openai/deployments/{1}/chat/completions?api-version={2}".format(
     config("AZURE_OPENAI_ENDPOINT", default=""),