Remove the old benchmark data

Author: Shashikant86

stackone_ai/models.py

@@ -584,7 +584,7 @@ def utility_tools(
 
         Utility tools enable dynamic tool discovery and execution based on natural language queries.
         By default, uses local hybrid BM25 + TF-IDF search. Optionally, can use cloud-based
-        semantic search for higher accuracy (84% Hit@5 vs 21% for local search).
+        semantic search for higher accuracy on natural language queries.
 
         Args:
             hybrid_alpha: Weight for BM25 in hybrid search (0-1). Only used when

stackone_ai/semantic_search.py

@@ -37,8 +37,7 @@ class SemanticSearchClient:
     """Client for StackOne semantic search API.
 
     This client provides access to the semantic search endpoint which uses
-    enhanced embeddings for 84% Hit@5 accuracy (compared to ~21% for local
-    BM25+TF-IDF search).
+    enhanced embeddings for higher accuracy than local BM25+TF-IDF search.
 
     Example:
         client = SemanticSearchClient(api_key="sk-xxx")

stackone_ai/toolset.py

@@ -296,9 +296,9 @@ def search_tools(
     ) -> Tools:
         """Search for and fetch tools using semantic search.
 
-        This method uses the StackOne semantic search API (84% Hit@5 accuracy)
-        to find relevant tools based on natural language queries. It optimizes
-        results by filtering to only connectors available in linked accounts.
+        This method uses the StackOne semantic search API to find relevant tools
+        based on natural language queries. It optimizes results by filtering to
+        only connectors available in linked accounts.
 
         Args:
             query: Natural language description of needed functionality

stackone_ai/utility_tools.py

@@ -270,8 +270,8 @@ def execute(
 def create_semantic_tool_search(semantic_client: SemanticSearchClient) -> StackOneTool:
     """Create a semantic search variant of tool_search.
 
-    Uses cloud semantic search API (84% Hit@5 accuracy) instead of
-    local BM25+TF-IDF (21% accuracy).
+    Uses cloud semantic search API instead of local BM25+TF-IDF for
+    improved natural language tool discovery.
 
     Args:
         semantic_client: Initialized SemanticSearchClient instance
@@ -287,7 +287,7 @@ def create_semantic_tool_search(semantic_client: SemanticSearchClient) -> StackO
     name = "tool_search"
     description = (
         "Searches for relevant tools based on a natural language query using "
-        "semantic vector search (84% accuracy). Call this first to discover "
+        "semantic vector search. Call this first to discover "
         "available tools before executing them."
     )
 

tests/benchmark_search.py

@@ -1,10 +1,7 @@
 """
 Benchmark comparing local BM25+TF-IDF vs semantic search.
 
-Expected results:
-- Local BM25+TF-IDF: ~21% Hit@5
-- Semantic Search: ~84% Hit@5
-- Improvement: 4x
+Compares Hit@5 and MRR between local BM25+TF-IDF and semantic search.
 
 Run with production API:
     STACKONE_API_KEY=xxx python tests/benchmark_search.py

tests/test_semantic_search.py

@@ -508,7 +508,6 @@ def test_semantic_tool_search_has_correct_parameters(self) -> None:
 
         assert tool.name == "tool_search"
         assert "semantic" in tool.description.lower()
-        assert "84%" in tool.description
 
         props = tool.parameters.properties
         assert "query" in props