update bert_model model_id

JaredforReal · JaredforReal · commit 2609c6b452e7 · 2025-10-16T14:38:20.000+08:00
Signed-off-by: JaredforReal &lt;w13431838023@gmail.com&gt;
diff --git a/config/config-mcp-classifier-example.yaml b/config/config-mcp-classifier-example.yaml
@@ -14,7 +14,7 @@
 
 # BERT model for semantic caching and tool selection
 bert_model:
-  model_id: "sentence-transformers/all-MiniLM-L6-v2"
+  model_id: models/all-MiniLM-L12-v2
   threshold: 0.85
   use_cpu: true
 
diff --git a/config/config.development.yaml b/config/config.development.yaml
@@ -3,7 +3,7 @@
 # for local development and debugging.
 
 bert_model:
-  model_id: sentence-transformers/all-MiniLM-L12-v2
+  model_id: models/all-MiniLM-L12-v2
   threshold: 0.6
   use_cpu: true
 
diff --git a/config/config.e2e.yaml b/config/config.e2e.yaml
@@ -1,5 +1,5 @@
 bert_model:
-  model_id: sentence-transformers/all-MiniLM-L12-v2
+  model_id: models/all-MiniLM-L12-v2
   threshold: 0.6
   use_cpu: true
 semantic_cache:
diff --git a/config/config.production.yaml b/config/config.production.yaml
@@ -3,7 +3,7 @@
 # for production deployment with Jaeger or other OTLP-compatible backends.
 
 bert_model:
-  model_id: sentence-transformers/all-MiniLM-L12-v2
+  model_id: models/all-MiniLM-L12-v2
   threshold: 0.6
   use_cpu: true
 
diff --git a/config/config.recipe-accuracy.yaml b/config/config.recipe-accuracy.yaml
@@ -13,7 +13,7 @@
 # - Jailbreak protection enabled
 
 bert_model:
-  model_id: sentence-transformers/all-MiniLM-L12-v2
+  model_id: models/all-MiniLM-L12-v2
   threshold: 0.7  # Higher threshold for better precision
   use_cpu: true
 
diff --git a/config/config.recipe-latency.yaml b/config/config.recipe-latency.yaml
@@ -13,7 +13,7 @@
 # - Minimal observability overhead
 
 bert_model:
-  model_id: sentence-transformers/all-MiniLM-L12-v2
+  model_id: models/all-MiniLM-L12-v2
   threshold: 0.4  # Very low threshold for fast matching
   use_cpu: true
 
diff --git a/config/config.recipe-token-efficiency.yaml b/config/config.recipe-token-efficiency.yaml
@@ -13,7 +13,7 @@
 # - Larger batch sizes for efficient processing
 
 bert_model:
-  model_id: sentence-transformers/all-MiniLM-L12-v2
+  model_id: models/all-MiniLM-L12-v2
   threshold: 0.5  # Lower threshold for faster matching
   use_cpu: true
 
diff --git a/config/config.testing.yaml b/config/config.testing.yaml
@@ -1,5 +1,5 @@
 bert_model:
-  model_id: sentence-transformers/all-MiniLM-L12-v2
+  model_id: models/all-MiniLM-L12-v2
   threshold: 0.6
   use_cpu: true
 
