Upload fine-tuned intent detection model - 2025-06-24 15:12:39

README.md

@@ -65,8 +65,8 @@ let model = try await LLMModelFactory.shared.loadContainer(
 ## Training Details
 
 - **Fine-tuning Method**: LoRA with model fusion
-- **Export Date**: 2025-06-24T14:59:43.091142
-- **Fusion Completed**: False
+- **Export Date**: 2025-06-24T15:12:36.041125
+- **Fusion Completed**: True
 
 ## Expected Outputs
 

model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:354f2dfb703806978bbb6839759f4c282df99ba54295b8b482e19545b2549cd4
+oid sha256:3c7d617ea9d68dc1d5205dee857d99bee5b1fb0be04d93f5abab20c7c810ed31
 size 868628547

model_config.json

@@ -1,13 +1,11 @@
 {
   "model_info": {
-    "model_id": "mlx-community/Qwen2.5-1.5B-Instruct-4bit",
+    "model_id": "knowledgebaseLLM",
     "model_type": "intent_detection",
     "version": "1.0",
-    "base_model": "mlx-community/Qwen2.5-1.5B-Instruct-4bit"
-  },
-  "adapter_files": {
-    "adapters": "adapters.safetensors",
-    "config": "adapter_config.json"
+    "base_model": "mlx-community/Qwen2.5-1.5B-Instruct-4bit",
+    "model_format": "complete_merged",
+    "adapter_merged": true
   },
   "inference_settings": {
     "max_tokens": 256,
@@ -47,11 +45,13 @@
     "examples": 5000,
     "iterations": 100,
     "training_completed": true,
-    "export_timestamp": "2025-06-24T14:59:43.091142"
+    "export_timestamp": "2025-06-24T15:12:36.041125",
+    "fusion_completed": true
   },
   "ios_integration": {
-    "adapter_path": "Model/adapters.safetensors",
-    "config_path": "Model/adapter_config.json",
+    "model_type": "complete_bundled",
+    "load_method": "local_bundle_only",
+    "requires_download": false,
     "system_prompt": "You are a helpful AI assistant for a knowledge-management app on an iPhone."
   }
 }