diff --git a/.github/workflows/e2e.yml b/.github/workflows/e2e.yml
index 2997c26f..cd366d30 100644
--- a/.github/workflows/e2e.yml
+++ b/.github/workflows/e2e.yml
@@ -44,7 +44,7 @@ jobs:
 
       - name: Run E2E tests
         env:
-          OPENAI_API_KEY: ${{ secrets.OPENAI_API_KEY }}
-          OPENAI_BASE_URL: ${{ secrets.OPENAI_BASE_URL }}
-          OPENAI_MODEL: ${{ secrets.OPENAI_MODEL }}
+          QWEN_CODE_API_KEY: ${{ secrets.OPENAI_API_KEY }}
+          QWEN_CODE_BASE_URL: ${{ secrets.OPENAI_BASE_URL }}
+          QWEN_CODE_MODEL: ${{ secrets.OPENAI_MODEL }}
         run: npm run test:integration:${{ matrix.sandbox }} -- --verbose --keep-output
diff --git a/README.md b/README.md
index a4c78c4d..d413f735 100644
--- a/README.md
+++ b/README.md
@@ -97,17 +97,17 @@ Qwen Code supports multiple API providers. You can configure your API key throug
 1. **Environment Variables**
 
    ```bash
-   export OPENAI_API_KEY="your_api_key_here"
-   export OPENAI_BASE_URL="your_api_endpoint"
-   export OPENAI_MODEL="your_model_choice"
+   export QWEN_CODE_API_KEY="your_api_key_here"
+   export QWEN_CODE_BASE_URL="your_api_endpoint"
+   export QWEN_CODE_MODEL="your_model_choice"
    ```
 
 2. **Project `.env` File**
    Create a `.env` file in your project root:
    ```env
-   OPENAI_API_KEY=your_api_key_here
-   OPENAI_BASE_URL=your_api_endpoint
-   OPENAI_MODEL=your_model_choice
+   QWEN_CODE_API_KEY=your_api_key_here
+   QWEN_CODE_BASE_URL=your_api_endpoint
+   QWEN_CODE_MODEL=your_model_choice
    ```
 
 #### API Provider Options
@@ -123,9 +123,9 @@ Qwen Code supports multiple API providers. You can configure your API key throug
 **Option 1: Alibaba Cloud Bailian** ([Apply for API Key](https://bailian.console.aliyun.com/))
 
 ```bash
-export OPENAI_API_KEY="your_api_key_here"
-export OPENAI_BASE_URL="https://dashscope.aliyuncs.com/compatible-mode/v1"
-export OPENAI_MODEL="qwen3-coder-plus"
+export QWEN_CODE_API_KEY="your_api_key_here"
+export QWEN_CODE_BASE_URL="https://dashscope.aliyuncs.com/compatible-mode/v1"
+export QWEN_CODE_MODEL="qwen3-coder-plus"
 ```
 
 **Option 2: ModelScope (Free Tier)** ([Apply for API Key](https://modelscope.cn/docs/model-service/API-Inference/intro))
@@ -134,9 +134,9 @@ export OPENAI_MODEL="qwen3-coder-plus"
 - ⚠️ Connect your Aliyun account to avoid authentication errors
 
 ```bash
-export OPENAI_API_KEY="your_api_key_here"
-export OPENAI_BASE_URL="https://api-inference.modelscope.cn/v1"
-export OPENAI_MODEL="Qwen/Qwen3-Coder-480B-A35B-Instruct"
+export QWEN_CODE_API_KEY="your_api_key_here"
+export QWEN_CODE_BASE_URL="https://api-inference.modelscope.cn/v1"
+export QWEN_CODE_MODEL="Qwen/Qwen3-Coder-480B-A35B-Instruct"
 ```
 
 </details>
@@ -147,17 +147,17 @@ export OPENAI_MODEL="Qwen/Qwen3-Coder-480B-A35B-Instruct"
 **Option 1: Alibaba Cloud ModelStudio** ([Apply for API Key](https://modelstudio.console.alibabacloud.com/))
 
 ```bash
-export OPENAI_API_KEY="your_api_key_here"
-export OPENAI_BASE_URL="https://dashscope-intl.aliyuncs.com/compatible-mode/v1"
-export OPENAI_MODEL="qwen3-coder-plus"
+export QWEN_CODE_API_KEY="your_api_key_here"
+export QWEN_CODE_BASE_URL="https://dashscope-intl.aliyuncs.com/compatible-mode/v1"
+export QWEN_CODE_MODEL="qwen3-coder-plus"
 ```
 
 **Option 2: OpenRouter (Free Tier Available)** ([Apply for API Key](https://openrouter.ai/))
 
 ```bash
-export OPENAI_API_KEY="your_api_key_here"
-export OPENAI_BASE_URL="https://openrouter.ai/api/v1"
-export OPENAI_MODEL="qwen/qwen3-coder:free"
+export QWEN_CODE_API_KEY="your_api_key_here"
+export QWEN_CODE_BASE_URL="https://openrouter.ai/api/v1"
+export QWEN_CODE_MODEL="qwen/qwen3-coder:free"
 ```
 
 </details>
diff --git a/docs/cli/configuration.md b/docs/cli/configuration.md
index 6b2cde3e..9b524ee2 100644
--- a/docs/cli/configuration.md
+++ b/docs/cli/configuration.md
@@ -219,8 +219,8 @@ In addition to a project settings file, a project's `.gemini` directory can cont
   - **Description:** Configures custom system prompt templates for specific model names and base URLs. This allows you to use different system prompts for different AI models or API endpoints.
   - **Default:** `undefined` (uses default system prompt)
   - **Properties:**
-    - **`baseUrls`** (array of strings, optional): Array of base URLs to exactly match against `OPENAI_BASE_URL` environment variable. If not specified, matches any base URL.
-    - **`modelNames`** (array of strings, optional): Array of model names to exactly match against `OPENAI_MODEL` environment variable. If not specified, matches any model.
+    - **`baseUrls`** (array of strings, optional): Array of base URLs to exactly match against `QWEN_CODE_BASE_URL` environment variable. If not specified, matches any base URL.
+    - **`modelNames`** (array of strings, optional): Array of model names to exactly match against `QWEN_CODE_MODEL` environment variable. If not specified, matches any model.
     - **`template`** (string): The system prompt template to use when both baseUrl and modelNames match. Supports placeholders:
       - `{RUNTIME_VARS_IS_GIT_REPO}`: Replaced with `true` or `false` based on whether the current directory is a git repository
       - `{RUNTIME_VARS_SANDBOX}`: Replaced with the sandbox type (e.g., `"sandbox-exec"`, `"docker"`, or empty string)
diff --git a/docs/cli/openai-auth.md b/docs/cli/openai-auth.md
index 9dd8c0ca..95f81e9e 100644
--- a/docs/cli/openai-auth.md
+++ b/docs/cli/openai-auth.md
@@ -40,9 +40,9 @@ qwen-code --openai-api-key "your-api-key-here" --model "gpt-4-turbo"
 Set the following environment variables in your shell or `.env` file:
 
 ```bash
-export OPENAI_API_KEY="your-api-key-here"
-export OPENAI_BASE_URL="https://api.openai.com/v1"  # Optional, defaults to this value
-export OPENAI_MODEL="gpt-4o"  # Optional, defaults to gpt-4o
+export QWEN_CODE_API_KEY="your-api-key-here"
+export QWEN_CODE_BASE_URL="https://api.openai.com/v1"  # Optional, defaults to this value
+export QWEN_CODE_MODEL="gpt-4o"  # Optional, defaults to gpt-4o
 ```
 
 ## Supported Models
@@ -58,7 +58,7 @@ The CLI supports all OpenAI models that are available through the OpenAI API, in
 
 ## Custom Endpoints
 
-You can use custom endpoints by setting the `OPENAI_BASE_URL` environment variable or using the `--openai-base-url` command line argument. This is useful for:
+You can use custom endpoints by setting the `QWEN_CODE_BASE_URL` environment variable or using the `--openai-base-url` command line argument. This is useful for:
 
 - Using Azure OpenAI
 - Using other OpenAI-compatible APIs
diff --git a/packages/cli/src/config/auth.ts b/packages/cli/src/config/auth.ts
index 45b44df0..10fbe4a8 100644
--- a/packages/cli/src/config/auth.ts
+++ b/packages/cli/src/config/auth.ts
@@ -39,8 +39,8 @@ export const validateAuthMethod = (authMethod: string): string | null => {
   }
 
   if (authMethod === AuthType.USE_OPENAI) {
-    if (!process.env.OPENAI_API_KEY) {
-      return 'OPENAI_API_KEY environment variable not found. You can enter it interactively or add it to your .env file.';
+    if (!process.env.QWEN_CODE_API_KEY) {
+      return 'QWEN_CODE_API_KEY environment variable not found. You can enter it interactively or add it to your .env file.';
     }
     return null;
   }
@@ -49,13 +49,13 @@ export const validateAuthMethod = (authMethod: string): string | null => {
 };
 
 export const setOpenAIApiKey = (apiKey: string): void => {
-  process.env.OPENAI_API_KEY = apiKey;
+  process.env.QWEN_CODE_API_KEY = apiKey;
 };
 
 export const setOpenAIBaseUrl = (baseUrl: string): void => {
-  process.env.OPENAI_BASE_URL = baseUrl;
+  process.env.QWEN_CODE_BASE_URL = baseUrl;
 };
 
 export const setOpenAIModel = (model: string): void => {
-  process.env.OPENAI_MODEL = model;
+  process.env.QWEN_CODE_MODEL = model;
 };
diff --git a/packages/cli/src/config/config.ts b/packages/cli/src/config/config.ts
index f3f9900a..e03f8768 100644
--- a/packages/cli/src/config/config.ts
+++ b/packages/cli/src/config/config.ts
@@ -264,12 +264,12 @@ export async function loadCliConfig(
 
   // Handle OpenAI API key from command line
   if (argv.openaiApiKey) {
-    process.env.OPENAI_API_KEY = argv.openaiApiKey;
+    process.env.QWEN_CODE_API_KEY = argv.openaiApiKey;
   }
 
   // Handle OpenAI base URL from command line
   if (argv.openaiBaseUrl) {
-    process.env.OPENAI_BASE_URL = argv.openaiBaseUrl;
+    process.env.QWEN_CODE_BASE_URL = argv.openaiBaseUrl;
   }
 
   // Set the context filename in the server's memoryTool module BEFORE loading memory
diff --git a/packages/cli/src/utils/sandbox.ts b/packages/cli/src/utils/sandbox.ts
index d7ea26d1..a726318f 100644
--- a/packages/cli/src/utils/sandbox.ts
+++ b/packages/cli/src/utils/sandbox.ts
@@ -517,15 +517,15 @@ export async function start_sandbox(
     args.push('--env', `GOOGLE_API_KEY=${process.env.GOOGLE_API_KEY}`);
   }
 
-  // copy OPENAI_API_KEY and related env vars for Qwen
-  if (process.env.OPENAI_API_KEY) {
-    args.push('--env', `OPENAI_API_KEY=${process.env.OPENAI_API_KEY}`);
+  // copy QWEN_CODE_API_KEY and related env vars for Qwen
+  if (process.env.QWEN_CODE_API_KEY) {
+    args.push('--env', `QWEN_CODE_API_KEY=${process.env.QWEN_CODE_API_KEY}`);
   }
-  if (process.env.OPENAI_BASE_URL) {
-    args.push('--env', `OPENAI_BASE_URL=${process.env.OPENAI_BASE_URL}`);
+  if (process.env.QWEN_CODE_BASE_URL) {
+    args.push('--env', `QWEN_CODE_BASE_URL=${process.env.QWEN_CODE_BASE_URL}`);
   }
-  if (process.env.OPENAI_MODEL) {
-    args.push('--env', `OPENAI_MODEL=${process.env.OPENAI_MODEL}`);
+  if (process.env.QWEN_CODE_MODEL) {
+    args.push('--env', `QWEN_CODE_MODEL=${process.env.QWEN_CODE_MODEL}`);
   }
 
   // copy GOOGLE_GENAI_USE_VERTEXAI
diff --git a/packages/core/src/config/models.ts b/packages/core/src/config/models.ts
index a9e5f745..d304f560 100644
--- a/packages/core/src/config/models.ts
+++ b/packages/core/src/config/models.ts
@@ -5,5 +5,5 @@
  */
 
 export const DEFAULT_GEMINI_MODEL = 'qwen3-coder-plus';
-export const DEFAULT_GEMINI_FLASH_MODEL = 'gemini-2.5-flash';
+export const DEFAULT_GEMINI_FLASH_MODEL = 'qwen3-coder-flash';
 export const DEFAULT_GEMINI_EMBEDDING_MODEL = 'gemini-embedding-001';
diff --git a/packages/core/src/core/__tests__/openaiTimeoutHandling.test.ts b/packages/core/src/core/__tests__/openaiTimeoutHandling.test.ts
index 81c77b61..b939f5b0 100644
--- a/packages/core/src/core/__tests__/openaiTimeoutHandling.test.ts
+++ b/packages/core/src/core/__tests__/openaiTimeoutHandling.test.ts
@@ -34,7 +34,7 @@ describe('OpenAIContentGenerator Timeout Handling', () => {
     vi.clearAllMocks();
 
     // Mock environment variables
-    vi.stubEnv('OPENAI_BASE_URL', '');
+    vi.stubEnv('QWEN_CODE_BASE_URL', '');
 
     // Mock config
     mockConfig = {
diff --git a/packages/core/src/core/contentGenerator.ts b/packages/core/src/core/contentGenerator.ts
index be1fe54a..2eba98ff 100644
--- a/packages/core/src/core/contentGenerator.ts
+++ b/packages/core/src/core/contentGenerator.ts
@@ -75,7 +75,7 @@ export async function createContentGeneratorConfig(
   const googleApiKey = process.env.GOOGLE_API_KEY || undefined;
   const googleCloudProject = process.env.GOOGLE_CLOUD_PROJECT || undefined;
   const googleCloudLocation = process.env.GOOGLE_CLOUD_LOCATION || undefined;
-  const openaiApiKey = process.env.OPENAI_API_KEY;
+  const openaiApiKey = process.env.QWEN_CODE_API_KEY;
 
   // Use runtime model from config if available, otherwise fallback to parameter or default
   const effectiveModel = model || DEFAULT_GEMINI_MODEL;
@@ -117,7 +117,7 @@ export async function createContentGeneratorConfig(
   if (authType === AuthType.USE_OPENAI && openaiApiKey) {
     contentGeneratorConfig.apiKey = openaiApiKey;
     contentGeneratorConfig.model =
-      process.env.OPENAI_MODEL || DEFAULT_GEMINI_MODEL;
+      process.env.QWEN_CODE_MODEL || DEFAULT_GEMINI_MODEL;
 
     return contentGeneratorConfig;
   }
diff --git a/packages/core/src/core/openaiContentGenerator.ts b/packages/core/src/core/openaiContentGenerator.ts
index 2df0c241..9e29ec5e 100644
--- a/packages/core/src/core/openaiContentGenerator.ts
+++ b/packages/core/src/core/openaiContentGenerator.ts
@@ -97,7 +97,7 @@ export class OpenAIContentGenerator implements ContentGenerator {
   constructor(apiKey: string, model: string, config: Config) {
     this.model = model;
     this.config = config;
-    const baseURL = process.env.OPENAI_BASE_URL || '';
+    const baseURL = process.env.QWEN_CODE_BASE_URL || '';
 
     // Configure timeout settings - using progressive timeouts
     const timeoutConfig = {
diff --git a/packages/core/src/core/prompts.test.ts b/packages/core/src/core/prompts.test.ts
index 66c206aa..7091b8e3 100644
--- a/packages/core/src/core/prompts.test.ts
+++ b/packages/core/src/core/prompts.test.ts
@@ -130,8 +130,8 @@ describe('URL matching with trailing slash compatibility', () => {
     // Test case 1: No trailing slash in config, actual URL has trailing slash
     process.env = {
       ...originalEnv,
-      OPENAI_BASE_URL: 'https://api.example.com/',
-      OPENAI_MODEL: 'gpt-4',
+      QWEN_CODE_BASE_URL: 'https://api.example.com/',
+      QWEN_CODE_MODEL: 'gpt-4',
     };
 
     const result1 = getCoreSystemPrompt(undefined, config);
@@ -140,8 +140,8 @@ describe('URL matching with trailing slash compatibility', () => {
     // Test case 2: Config has trailing slash, actual URL has no trailing slash
     process.env = {
       ...originalEnv,
-      OPENAI_BASE_URL: 'https://api.openai.com',
-      OPENAI_MODEL: 'gpt-3.5-turbo',
+      QWEN_CODE_BASE_URL: 'https://api.openai.com',
+      QWEN_CODE_MODEL: 'gpt-3.5-turbo',
     };
 
     const result2 = getCoreSystemPrompt(undefined, config);
@@ -150,8 +150,8 @@ describe('URL matching with trailing slash compatibility', () => {
     // Test case 3: No trailing slash in config, actual URL has no trailing slash
     process.env = {
       ...originalEnv,
-      OPENAI_BASE_URL: 'https://api.example.com',
-      OPENAI_MODEL: 'gpt-4',
+      QWEN_CODE_BASE_URL: 'https://api.example.com',
+      QWEN_CODE_MODEL: 'gpt-4',
     };
 
     const result3 = getCoreSystemPrompt(undefined, config);
@@ -160,8 +160,8 @@ describe('URL matching with trailing slash compatibility', () => {
     // Test case 4: Config has trailing slash, actual URL has trailing slash
     process.env = {
       ...originalEnv,
-      OPENAI_BASE_URL: 'https://api.openai.com/',
-      OPENAI_MODEL: 'gpt-3.5-turbo',
+      QWEN_CODE_BASE_URL: 'https://api.openai.com/',
+      QWEN_CODE_MODEL: 'gpt-3.5-turbo',
     };
 
     const result4 = getCoreSystemPrompt(undefined, config);
@@ -187,8 +187,8 @@ describe('URL matching with trailing slash compatibility', () => {
     // Test case: URLs do not match
     process.env = {
       ...originalEnv,
-      OPENAI_BASE_URL: 'https://api.different.com',
-      OPENAI_MODEL: 'gpt-4',
+      QWEN_CODE_BASE_URL: 'https://api.different.com',
+      QWEN_CODE_MODEL: 'gpt-4',
     };
 
     const result = getCoreSystemPrompt(undefined, config);
diff --git a/packages/core/src/core/prompts.ts b/packages/core/src/core/prompts.ts
index 7a397455..372c5719 100644
--- a/packages/core/src/core/prompts.ts
+++ b/packages/core/src/core/prompts.ts
@@ -65,8 +65,8 @@ export function getCoreSystemPrompt(
 
   // Check for system prompt mappings from global config
   if (config?.systemPromptMappings) {
-    const currentModel = process.env.OPENAI_MODEL || DEFAULT_GEMINI_MODEL;
-    const currentBaseUrl = process.env.OPENAI_BASE_URL || '';
+    const currentModel = process.env.QWEN_CODE_MODEL || DEFAULT_GEMINI_MODEL;
+    const currentBaseUrl = process.env.QWEN_CODE_BASE_URL || '';
 
     const matchedMapping = config.systemPromptMappings.find((mapping) => {
       const { baseUrls, modelNames } = mapping;