diff --git a/README.md b/README.md
index 2015277..99b0186 100644
--- a/README.md
+++ b/README.md
@@ -81,7 +81,7 @@ Diagrams are represented as XML that can be rendered in draw.io. The AI processe
 ## Multi-Provider Support
 
 -   AWS Bedrock (default)
--   OpenAI / OpenAI-compatible APIs (via `OPENAI_BASE_URL`)
+-   OpenAI
 -   Anthropic
 -   Google AI
 -   Azure OpenAI
@@ -89,6 +89,12 @@ Diagrams are represented as XML that can be rendered in draw.io. The AI processe
 -   OpenRouter
 -   DeepSeek
 
+All providers except AWS Bedrock and OpenRouter support custom endpoints.
+
+📖 **[Detailed Provider Configuration Guide](./docs/ai-providers.md)** - See setup instructions for each provider.
+
+**Model Requirements**: This task requires strong model capabilities for generating long-form text with strict formatting constraints (draw.io XML). Recommended models include Claude Sonnet 4.5, GPT-4o, Gemini 2.0, and DeepSeek V3/R1.
+
 Note that `claude-sonnet-4-5` has trained on draw.io diagrams with AWS logos, so if you want to create AWS architecture diagrams, this is the best choice.
 
 ## Getting Started
@@ -144,7 +150,7 @@ Edit `.env.local` and configure your chosen provider:
 -   Set `AI_MODEL` to the specific model you want to use
 -   Add the required API keys for your provider
 
-See the [Multi-Provider Support](#multi-provider-support) section above for provider-specific configuration examples.
+See the [Provider Configuration Guide](./docs/ai-providers.md) for detailed setup instructions for each provider.
 
 4. Run the development server:
 
diff --git a/README_CN.md b/README_CN.md
index fa96a28..c7f9914 100644
--- a/README_CN.md
+++ b/README_CN.md
@@ -81,7 +81,7 @@ https://github.com/user-attachments/assets/b2eef5f3-b335-4e71-a755-dc2e80931979
 ## 多提供商支持
 
 -   AWS Bedrock（默认）
--   OpenAI / OpenAI兼容API（通过 `OPENAI_BASE_URL`）
+-   OpenAI
 -   Anthropic
 -   Google AI
 -   Azure OpenAI
@@ -89,6 +89,12 @@ https://github.com/user-attachments/assets/b2eef5f3-b335-4e71-a755-dc2e80931979
 -   OpenRouter
 -   DeepSeek
 
+除AWS Bedrock和OpenRouter外，所有提供商都支持自定义端点。
+
+📖 **[详细的提供商配置指南](./docs/ai-providers.md)** - 查看各提供商的设置说明。
+
+**模型要求**：此任务需要强大的模型能力，因为它涉及生成具有严格格式约束的长文本（draw.io XML）。推荐使用Claude Sonnet 4.5、GPT-4o、Gemini 2.0和DeepSeek V3/R1。
+
 注意：`claude-sonnet-4-5` 已在带有AWS标志的draw.io图表上进行训练，因此如果您想创建AWS架构图，这是最佳选择。
 
 ## 快速开始
@@ -144,7 +150,7 @@ cp env.example .env.local
 -   将 `AI_MODEL` 设置为您要使用的特定模型
 -   添加您的提供商所需的API密钥
 
-请参阅上面的[多提供商支持](#多提供商支持)部分了解特定提供商的配置示例。
+详细设置说明请参阅[提供商配置指南](./docs/ai-providers.md)。
 
 4. 运行开发服务器：
 
diff --git a/README_JA.md b/README_JA.md
index a61195d..ac83cdc 100644
--- a/README_JA.md
+++ b/README_JA.md
@@ -81,7 +81,7 @@ https://github.com/user-attachments/assets/b2eef5f3-b335-4e71-a755-dc2e80931979
 ## マルチプロバイダーサポート
 
 -   AWS Bedrock（デフォルト）
--   OpenAI / OpenAI互換API（`OPENAI_BASE_URL`経由）
+-   OpenAI
 -   Anthropic
 -   Google AI
 -   Azure OpenAI
@@ -89,6 +89,12 @@ https://github.com/user-attachments/assets/b2eef5f3-b335-4e71-a755-dc2e80931979
 -   OpenRouter
 -   DeepSeek
 
+AWS BedrockとOpenRouter以外のすべてのプロバイダーはカスタムエンドポイントをサポートしています。
+
+📖 **[詳細なプロバイダー設定ガイド](./docs/ai-providers.md)** - 各プロバイダーの設定手順をご覧ください。
+
+**モデル要件**：このタスクは厳密なフォーマット制約（draw.io XML）を持つ長文テキスト生成を伴うため、強力なモデル機能が必要です。Claude Sonnet 4.5、GPT-4o、Gemini 2.0、DeepSeek V3/R1を推奨します。
+
 注：`claude-sonnet-4-5`はAWSロゴ付きのdraw.ioダイアグラムで学習されているため、AWSアーキテクチャダイアグラムを作成したい場合は最適な選択です。
 
 ## はじめに
@@ -144,7 +150,7 @@ cp env.example .env.local
 -   `AI_MODEL`を使用する特定のモデルに設定
 -   プロバイダーに必要なAPIキーを追加
 
-プロバイダー固有の設定例については、上記の[マルチプロバイダーサポート](#マルチプロバイダーサポート)セクションを参照してください。
+詳細な設定手順については[プロバイダー設定ガイド](./docs/ai-providers.md)を参照してください。
 
 4. 開発サーバーを起動：
 
diff --git a/docs/ai-providers.md b/docs/ai-providers.md
new file mode 100644
index 0000000..aabae67
--- /dev/null
+++ b/docs/ai-providers.md
@@ -0,0 +1,141 @@
+# AI Provider Configuration
+
+This guide explains how to configure different AI model providers for next-ai-draw-io.
+
+## Quick Start
+
+1. Copy `.env.example` to `.env.local`
+2. Set your API key for your chosen provider
+3. Set `AI_MODEL` to your desired model
+4. Run `npm run dev`
+
+## Supported Providers
+
+### Google Gemini
+
+```bash
+GOOGLE_GENERATIVE_AI_API_KEY=your_api_key
+AI_MODEL=gemini-2.0-flash
+```
+
+Optional custom endpoint:
+
+```bash
+GOOGLE_BASE_URL=https://your-custom-endpoint
+```
+
+### OpenAI
+
+```bash
+OPENAI_API_KEY=your_api_key
+AI_MODEL=gpt-4o
+```
+
+Optional custom endpoint (for OpenAI-compatible services):
+
+```bash
+OPENAI_BASE_URL=https://your-custom-endpoint/v1
+```
+
+### Anthropic
+
+```bash
+ANTHROPIC_API_KEY=your_api_key
+AI_MODEL=claude-sonnet-4-5-20250514
+```
+
+Optional custom endpoint:
+
+```bash
+ANTHROPIC_BASE_URL=https://your-custom-endpoint
+```
+
+### DeepSeek
+
+```bash
+DEEPSEEK_API_KEY=your_api_key
+AI_MODEL=deepseek-chat
+```
+
+Optional custom endpoint:
+
+```bash
+DEEPSEEK_BASE_URL=https://your-custom-endpoint
+```
+
+### Azure OpenAI
+
+```bash
+AZURE_API_KEY=your_api_key
+AI_MODEL=your-deployment-name
+```
+
+Optional custom endpoint:
+
+```bash
+AZURE_BASE_URL=https://your-resource.openai.azure.com
+```
+
+### AWS Bedrock
+
+```bash
+AWS_REGION=us-west-2
+AWS_ACCESS_KEY_ID=your_access_key_id
+AWS_SECRET_ACCESS_KEY=your_secret_access_key
+AI_MODEL=anthropic.claude-sonnet-4-5-20250514-v1:0
+```
+
+Note: On AWS (Amplify, Lambda, EC2 with IAM role), credentials are automatically obtained from the IAM role.
+
+### OpenRouter
+
+```bash
+OPENROUTER_API_KEY=your_api_key
+AI_MODEL=anthropic/claude-sonnet-4
+```
+
+Optional custom endpoint:
+
+```bash
+OPENROUTER_BASE_URL=https://your-custom-endpoint
+```
+
+### Ollama (Local)
+
+```bash
+AI_PROVIDER=ollama
+AI_MODEL=llama3.2
+```
+
+Optional custom URL:
+
+```bash
+OLLAMA_BASE_URL=http://localhost:11434
+```
+
+## Auto-Detection
+
+If you only configure **one** provider's API key, the system will automatically detect and use that provider. No need to set `AI_PROVIDER`.
+
+If you configure **multiple** API keys, you must explicitly set `AI_PROVIDER`:
+
+```bash
+AI_PROVIDER=google  # or: openai, anthropic, deepseek, azure, bedrock, openrouter, ollama
+```
+
+## Model Capability Requirements
+
+This task requires exceptionally strong model capabilities, as it involves generating long-form text with strict formatting constraints (draw.io XML).
+
+**Recommended models**:
+
+-   Claude Sonnet 4.5 / Opus 4.5
+
+**Note on Ollama**: While Ollama is supported as a provider, it's generally not practical for this use case unless you're running high-capability models like DeepSeek R1 or Qwen3-235B locally.
+
+## Recommendations
+
+-   **Best experience**: Use models with vision support (GPT-4o, Claude, Gemini) for image-to-diagram features
+-   **Budget-friendly**: DeepSeek offers competitive pricing
+-   **Privacy**: Use Ollama for fully local, offline operation (requires powerful hardware)
+-   **Flexibility**: OpenRouter provides access to many models through a single API