bug

tabs/run_benchmark.py

@@ -23,9 +23,9 @@ def run_benchmark_main(tool_name, model_name, num_questions, openai_api_key, ant
     if openrouter_api_key:
         kwargs["api_keys"]["openrouter"] = openrouter_api_key
 
-    if "gpt" in model:
+    if "gpt" in model_name:
         kwargs["llm_provider"] = "openai"
-    elif "claude" in model:
+    elif "claude" in model_name:
         kwargs["llm_provider"] = "anthropic"
     else:
         kwargs["llm_provider"] = "openrouter"