From c0d0e13b2db64b57b2fbdfeba663e780bea600ef Mon Sep 17 00:00:00 2001
From: Roo Code <roomote@roocode.com>
Date: Tue, 4 Nov 2025 10:41:10 +0000
Subject: [PATCH] docs: add explanatory comment for GLM-4.6 thinking parameter

Added comment explaining why GLM-4.6 uses thinking parameter instead of reasoning_effort for better maintainability
---
 src/api/providers/openai.ts | 2 ++
 1 file changed, 2 insertions(+)

diff --git a/src/api/providers/openai.ts b/src/api/providers/openai.ts
index fc62d6dae9..9b01e3c3e2 100644
--- a/src/api/providers/openai.ts
+++ b/src/api/providers/openai.ts
@@ -95,6 +95,8 @@ export class OpenAiHandler extends BaseProvider implements SingleCompletionHandl
 		const ark = modelUrl.includes(".volces.com")
 
 		// Check if this is GLM-4.6 model with reasoning support
+		// GLM-4.6 uses the 'thinking' parameter instead of 'reasoning_effort' for enabling reasoning
+		// This is a vendor-specific implementation detail for Z AI's GLM models
 		const isGLM46WithReasoning =
 			modelId.includes("glm-4.6") &&
 			this.options.enableReasoningEffort &&