feat: 添加 XMind 模型供应商支持

- 添加 XMind OpenAI 兼容模型（11个模型） - 添加 XMind Anthropic 兼容模型（5个模型） - 添加 XMind Vertex AI 兼容模型（3个模型，使用 OpenAI 格式） - 更新 deploy.sh 支持 XMind API 密钥环境变量 - 添加 build-and-push.sh 脚本用于构建和推送 Docker 镜像 - 更新 README.md 文档说明
2026-01-13 08:49:39 +00:00
parent 120b395863
commit d25f43e25b
4 changed files with 336 additions and 22 deletions
@@ -1,6 +1,6 @@
 # LiteLLM Azure Container Apps 部署

-本项目用于在 Azure Container Apps 上部署 LiteLLM 代理服务器，支持 Taiji AI Cloud 和 OpenRouter 两个模型供应商。
+本项目用于在 Azure Container Apps 上部署 LiteLLM 代理服务器，支持 Taiji AI Cloud、OpenRouter 和 XMind 三个模型供应商。

 ## 架构概览

@@ -25,12 +25,13 @@
 │  Cache (SSL)    │  │    (SSL)        │  │                 │
 │  Port 10000     │  │    Port 5432    │  │ • Taiji AI Cloud│
 └─────────────────┘  └─────────────────┘  │ • OpenRouter    │
+                                          │ • XMind         │
                                          └─────────────────┘
 ```

 ## 功能特性

- ✅ **多模型供应商支持**: Taiji AI Cloud 和 OpenRouter
+- ✅ **多模型供应商支持**: Taiji AI Cloud、OpenRouter 和 XMind
 - ✅ **通配符路由**: 支持 `taiji/*` 和 `openrouter/*` 动态模型路由
 - ✅ **Redis SSL 连接**: 安全连接 Azure Redis Cache
 - ✅ **PostgreSQL 数据库**: 持久化存储 API 密钥和使用记录
@@ -67,6 +68,21 @@ vim .env

 ### 2. 构建并推送镜像

+#### 方法一：使用快速脚本（推荐）
+
+```bash
+# 赋予脚本执行权限
+chmod +x build-and-push.sh
+
+# 构建并推送到 ACR（使用 latest 标签）
+./build-and-push.sh
+
+# 或指定标签
+./build-and-push.sh v1.0.0
+```
+
+#### 方法二：使用完整部署脚本
+
 ```bash
 # 赋予脚本执行权限
 chmod +x deploy.sh
@@ -140,6 +156,37 @@ curl -X POST https://litellm.graystone-fb459c5d.southeastasia.azurecontainerapps
  }'
 ```

+#### 使用 XMind 模型
+
+```bash
+# OpenAI 兼容模型
+curl -X POST https://litellm.graystone-fb459c5d.southeastasia.azurecontainerapps.io/chat/completions \
+  -H "Authorization: Bearer sk-litellm-master-key" \
+  -H "Content-Type: application/json" \
+  -d '{
+    "model": "xmind/gpt-4o",
+    "messages": [{"role": "user", "content": "Hello!"}]
+  }'
+
+# Anthropic 兼容模型
+curl -X POST https://litellm.graystone-fb459c5d.southeastasia.azurecontainerapps.io/chat/completions \
+  -H "Authorization: Bearer sk-litellm-master-key" \
+  -H "Content-Type: application/json" \
+  -d '{
+    "model": "xmind/claude-3.7-sonnet",
+    "messages": [{"role": "user", "content": "Hello!"}]
+  }'
+
+# Vertex AI 兼容模型
+curl -X POST https://litellm.graystone-fb459c5d.southeastasia.azurecontainerapps.io/chat/completions \
+  -H "Authorization: Bearer sk-litellm-master-key" \
+  -H "Content-Type: application/json" \
+  -d '{
+    "model": "xmind/gemini-2.5-pro",
+    "messages": [{"role": "user", "content": "Hello!"}]
+  }'
+```
+
 ### 环境变量说明

 | 变量 | 说明 | 必需 |
@@ -153,6 +200,10 @@ curl -X POST https://litellm.graystone-fb459c5d.southeastasia.azurecontainerapps
 | `REDIS_SSL` | 启用 Redis SSL（true/false） | ✅ |
 | `TAIJI_API_KEY` | Taiji AI Cloud API 密钥 | ✅ |
 | `OPENROUTER_API_KEY` | OpenRouter API 密钥 | ✅ |
+| `XMIND_API_KEY_OPENAI` | XMind OpenAI 兼容 API 密钥（有默认值） | ❌ |
+| `XMIND_API_KEY_ANTHROPIC` | XMind Anthropic 兼容 API 密钥（有默认值） | ❌ |
+| `XMIND_API_KEY_VERTEX_AI` | XMind Vertex AI 兼容 API 密钥（有默认值） | ❌ |
+| `ACR_PASSWORD` | Azure Container Registry 密码（用于推送镜像） | ✅ |
 | `UI_USERNAME` | Admin UI 用户名 | ❌ |
 | `UI_PASSWORD` | Admin UI 密码 | ❌ |

@@ -238,7 +289,8 @@ litellm/
 ├── config.yaml          # LiteLLM 主配置文件
 ├── Dockerfile           # Docker 构建文件
 ├── .env.example         # 环境变量示例
-├── deploy.sh            # 部署脚本
+├── deploy.sh            # 完整部署脚本（构建、推送、部署）
+├── build-and-push.sh    # 快速构建并推送镜像脚本
 └── README.md            # 本文档
 ```

@@ -0,0 +1,64 @@
+#!/bin/bash
+# 快速构建并推送镜像到 Azure Container Registry
+# 使用方法: ./build-and-push.sh [tag]
+
+set -e
+
+# 加载环境变量
+if [ -f .env ]; then
+    export $(cat .env | grep -v '^#' | xargs)
+fi
+
+# 配置
+ACR_SERVER="${ACR_SERVER:-agnettaiji.azurecr.io}"
+ACR_USERNAME="${ACR_USERNAME:-agnettaiji}"
+IMAGE_NAME="${IMAGE_NAME:-litellm}"
+
+# 如果没有提供标签，使用时间戳格式: vYYYYMMDDHHMMSS
+if [ -z "$1" ]; then
+    IMAGE_TAG="v$(date +%Y%m%d%H%M%S)"
+else
+    IMAGE_TAG="$1"
+fi
+
+FULL_IMAGE_NAME="${ACR_SERVER}/${IMAGE_NAME}:${IMAGE_TAG}"
+
+echo "=========================================="
+echo "构建并推送 Docker 镜像到 ACR"
+echo "=========================================="
+echo "镜像名称: ${FULL_IMAGE_NAME}"
+echo ""
+
+# 检查 ACR 密码
+if [ -z "$ACR_PASSWORD" ]; then
+    echo "错误: 未设置 ACR_PASSWORD 环境变量"
+    echo "请设置 ACR_PASSWORD 或创建 .env 文件"
+    exit 1
+fi
+
+# 构建镜像（同时打 latest 标签，方便快速引用）
+echo "步骤 1/2: 构建 Docker 镜像..."
+docker build -t "${FULL_IMAGE_NAME}" -t "${ACR_SERVER}/${IMAGE_NAME}:latest" .
+
+# 登录 ACR
+echo ""
+echo "步骤 2/2: 推送镜像到 ACR..."
+echo "${ACR_PASSWORD}" | docker login "${ACR_SERVER}" -u "${ACR_USERNAME}" --password-stdin
+
+# 推送镜像（推送带版本号的标签和 latest 标签）
+docker push "${FULL_IMAGE_NAME}"
+docker push "${ACR_SERVER}/${IMAGE_NAME}:latest"
+
+echo ""
+echo "=========================================="
+echo "完成！镜像已推送到: ${FULL_IMAGE_NAME}"
+echo "同时更新了 latest 标签: ${ACR_SERVER}/${IMAGE_NAME}:latest"
+echo "=========================================="
+echo ""
+echo "镜像标签: ${IMAGE_TAG}"
+echo ""
+echo "下一步: 运行以下命令部署应用:"
+echo "  ./deploy.sh update"
+echo ""
+echo "或者运行完整部署:"
+echo "  ./deploy.sh deploy"
@@ -25,6 +25,37 @@ general_settings:
  
  # VPC 内允许请求继续处理（即使数据库不可用）
  allow_requests_on_db_unavailable: true
+  
+  # ==========================================================================
+  # 健康检查设置 - 完全禁用
+  # ==========================================================================
+  # 禁用所有后台健康检查（不对模型供应商发起健康检测请求）
+  background_health_checks: false
+  # 健康检查间隔设置为 0 完全禁用
+  health_check_interval: 0
+
+# =============================================================================
+# 回调设置 - 计费 Webhook
+# =============================================================================
+litellm_settings:
+  # 启用回调 - 使用 custom_callback_api 发送到 GENERIC_LOGGER_ENDPOINT
+  success_callback: ["custom_callback_api"]
+  failure_callback: ["custom_callback_api"]
+  
+  # 请求超时（秒）
+  request_timeout: 600
+  
+  # 日志设置
+  set_verbose: false
+  json_logs: true
+  
+  # 缓存配置
+  cache: true
+  cache_params:
+    type: redis
+    host: os.environ/REDIS_HOST
+    port: os.environ/REDIS_PORT
+    password: os.environ/REDIS_PASSWORD

 # =============================================================================
 # 模型列表配置
@@ -4491,6 +4522,178 @@ model_list:
    model_info:
      disable_background_health_check: true

+  # -------------------------------------------------------------------------
+  # XMind 模型（OpenAI 兼容）
+  # 总计: 12 个模型
+  # -------------------------------------------------------------------------
+  - model_name: xmind/gpt-4o-mini
+    litellm_params:
+      model: openai/gpt-4o-mini
+      api_base: https://apis.openroutex.com/oai/v1
+      api_key: os.environ/XMIND_API_KEY_OPENAI
+    model_info:
+      disable_background_health_check: true
+
+  - model_name: xmind/gpt-4o
+    litellm_params:
+      model: openai/gpt-4o
+      api_base: https://apis.openroutex.com/oai/v1
+      api_key: os.environ/XMIND_API_KEY_OPENAI
+    model_info:
+      disable_background_health_check: true
+
+  - model_name: xmind/gpt-4.1
+    litellm_params:
+      model: openai/gpt-4.1
+      api_base: https://apis.openroutex.com/oai/v1
+      api_key: os.environ/XMIND_API_KEY_OPENAI
+    model_info:
+      disable_background_health_check: true
+
+  - model_name: xmind/gpt-image-1
+    litellm_params:
+      model: openai/gpt-image-1
+      api_base: https://apis.openroutex.com/oai/v1
+      api_key: os.environ/XMIND_API_KEY_OPENAI
+    model_info:
+      disable_background_health_check: true
+
+  - model_name: xmind/gpt-5-nano
+    litellm_params:
+      model: openai/gpt-5-nano
+      api_base: https://apis.openroutex.com/oai/v1
+      api_key: os.environ/XMIND_API_KEY_OPENAI
+    model_info:
+      disable_background_health_check: true
+
+  - model_name: xmind/gpt-5-mini
+    litellm_params:
+      model: openai/gpt-5-mini
+      api_base: https://apis.openroutex.com/oai/v1
+      api_key: os.environ/XMIND_API_KEY_OPENAI
+    model_info:
+      disable_background_health_check: true
+
+  - model_name: xmind/gpt-5
+    litellm_params:
+      model: openai/gpt-5
+      api_base: https://apis.openroutex.com/oai/v1
+      api_key: os.environ/XMIND_API_KEY_OPENAI
+    model_info:
+      disable_background_health_check: true
+
+  - model_name: xmind/gpt-5.1-chat
+    litellm_params:
+      model: openai/gpt-5.1-chat
+      api_base: https://apis.openroutex.com/oai/v1
+      api_key: os.environ/XMIND_API_KEY_OPENAI
+    model_info:
+      disable_background_health_check: true
+
+  - model_name: xmind/gpt-5.1
+    litellm_params:
+      model: openai/gpt-5.1
+      api_base: https://apis.openroutex.com/oai/v1
+      api_key: os.environ/XMIND_API_KEY_OPENAI
+    model_info:
+      disable_background_health_check: true
+
+  - model_name: xmind/gpt-5.2
+    litellm_params:
+      model: openai/gpt-5.2
+      api_base: https://apis.openroutex.com/oai/v1
+      api_key: os.environ/XMIND_API_KEY_OPENAI
+    model_info:
+      disable_background_health_check: true
+
+  - model_name: xmind/gpt-5.2-chat
+    litellm_params:
+      model: openai/gpt-5.2-chat
+      api_base: https://apis.openroutex.com/oai/v1
+      api_key: os.environ/XMIND_API_KEY_OPENAI
+    model_info:
+      disable_background_health_check: true
+
+  # -------------------------------------------------------------------------
+  # XMind 模型（Anthropic 兼容）
+  # 总计: 5 个模型
+  # 端点: https://anthropic.openroutex.com/v1/messages
+  # 请求头: x-api-key (API 密钥), anthropic-version (API 版本，例如 2023-06-01)
+  # 注意: 使用 anthropic provider，api_base 设置为完整端点路径
+  # 模型名称必须使用完整的模型 ID，包括日期后缀
+  # -------------------------------------------------------------------------
+  - model_name: xmind/claude-3.7-sonnet
+    litellm_params:
+      model: anthropic/claude-3-7-sonnet-20250219
+      api_base: https://anthropic.openroutex.com/v1/messages
+      api_key: os.environ/XMIND_API_KEY_ANTHROPIC
+    model_info:
+      disable_background_health_check: true
+
+  - model_name: xmind/claude-opus-4.1
+    litellm_params:
+      model: anthropic/claude-opus-4-1-20250805
+      api_base: https://anthropic.openroutex.com/v1/messages
+      api_key: os.environ/XMIND_API_KEY_ANTHROPIC
+    model_info:
+      disable_background_health_check: true
+
+  - model_name: xmind/claude-sonnet-4.5
+    litellm_params:
+      model: anthropic/claude-sonnet-4-5-20250929
+      api_base: https://anthropic.openroutex.com/v1/messages
+      api_key: os.environ/XMIND_API_KEY_ANTHROPIC
+    model_info:
+      disable_background_health_check: true
+
+  - model_name: xmind/claude-haiku-4.5
+    litellm_params:
+      model: anthropic/claude-haiku-4-5-20251001
+      api_base: https://anthropic.openroutex.com/v1/messages
+      api_key: os.environ/XMIND_API_KEY_ANTHROPIC
+    model_info:
+      disable_background_health_check: true
+
+  - model_name: xmind/claude-opus-4.5
+    litellm_params:
+      model: anthropic/claude-opus-4-5-20251101
+      api_base: https://anthropic.openroutex.com/v1/messages
+      api_key: os.environ/XMIND_API_KEY_ANTHROPIC
+    model_info:
+      disable_background_health_check: true
+
+  # -------------------------------------------------------------------------
+  # XMind 模型（Google Vertex AI 兼容 - OpenAI 格式）
+  # 总计: 3 个模型
+  # 端点: https://vertexai.openroutex.com/oai/v1/chat/completions
+  # 请求头: Authorization: Bearer {api_key}
+  # 注意: 使用标准的 OpenAI 兼容 API 格式，使用 openai provider
+  # 模型名称直接使用模型 ID（如 gemini-2.5-pro）
+  # -------------------------------------------------------------------------
+  - model_name: xmind/gemini-2.5-pro
+    litellm_params:
+      model: openai/gemini-2.5-pro
+      api_base: https://vertexai.openroutex.com/oai/v1
+      api_key: os.environ/XMIND_API_KEY_VERTEX_AI
+    model_info:
+      disable_background_health_check: true
+
+  - model_name: xmind/gemini-2.5-flash
+    litellm_params:
+      model: openai/gemini-2.5-flash
+      api_base: https://vertexai.openroutex.com/oai/v1
+      api_key: os.environ/XMIND_API_KEY_VERTEX_AI
+    model_info:
+      disable_background_health_check: true
+
+  - model_name: xmind/gemini-3-pro-preview
+    litellm_params:
+      model: openai/gemini-3-pro-preview
+      api_base: https://vertexai.openroutex.com/oai/v1
+      api_key: os.environ/XMIND_API_KEY_VERTEX_AI
+    model_info:
+      disable_background_health_check: true
+
 # =============================================================================
 # 路由器设置
 # =============================================================================
@@ -4508,25 +4711,6 @@ router_settings:
  cooldown_time: 30
  retry_after: 10

-# =============================================================================
-# LiteLLM 设置
-# =============================================================================
-litellm_settings:
-  # 请求超时（秒）
-  request_timeout: 600
-  
-  # 日志设置
-  set_verbose: false
-  json_logs: true
-  
-  # 缓存配置
-  cache: true
-  cache_params:
-    type: redis
-    host: os.environ/REDIS_HOST
-    port: os.environ/REDIS_PORT
-    password: os.environ/REDIS_PASSWORD
-
 # =============================================================================
 # 环境变量说明
 # =============================================================================
@@ -4540,3 +4724,6 @@ litellm_settings:
 # - REDIS_SSL: 是否启用 Redis SSL（true/false）
 # - TAIJI_API_KEY: Taiji AI Cloud API 密钥
 # - OPENROUTER_API_KEY: OpenRouter API 密钥
+# - XMIND_API_KEY_OPENAI: XMind OpenAI 兼容 API 密钥（默认: 6b761f38b4bf4c829390d82a43f8f32f）
+# - XMIND_API_KEY_ANTHROPIC: XMind Anthropic 兼容 API 密钥（默认: 9b123e68b3734856a6983fa229a1f94d）
+# - XMIND_API_KEY_VERTEX_AI: XMind Vertex AI 兼容 API 密钥（默认: 854a5b589db44fb0a08c988dd2dd57f8）
@@ -26,6 +26,11 @@ IMAGE_TAG="${IMAGE_TAG:-latest}"
 # 完整镜像名称
 FULL_IMAGE_NAME="${ACR_SERVER}/${IMAGE_NAME}:${IMAGE_TAG}"

+# XMind API 密钥（可选，有默认值）
+XMIND_API_KEY_OPENAI="${XMIND_API_KEY_OPENAI:-6b761f38b4bf4c829390d82a43f8f32f}"
+XMIND_API_KEY_ANTHROPIC="${XMIND_API_KEY_ANTHROPIC:-9b123e68b3734856a6983fa229a1f94d}"
+XMIND_API_KEY_VERTEX_AI="${XMIND_API_KEY_VERTEX_AI:-854a5b589db44fb0a08c988dd2dd57f8}"
+
 # =============================================================================
 # 辅助函数
 # =============================================================================
@@ -102,6 +107,9 @@ cmd_update() {
            "REDIS_SSL=true" \
            "TAIJI_API_KEY=secretref:taiji-api-key" \
            "OPENROUTER_API_KEY=secretref:openrouter-api-key" \
+            "XMIND_API_KEY_OPENAI=secretref:xmind-api-key-openai" \
+            "XMIND_API_KEY_ANTHROPIC=secretref:xmind-api-key-anthropic" \
+            "XMIND_API_KEY_VERTEX_AI=secretref:xmind-api-key-vertex-ai" \
            "LITELLM_MODE=PRODUCTION" \
            "LITELLM_LOG=ERROR" \
            "UI_USERNAME=${UI_USERNAME:-admin}" \
@@ -127,6 +135,9 @@ cmd_secrets() {
            "redis-password=${REDIS_PASSWORD}" \
            "taiji-api-key=${TAIJI_API_KEY}" \
            "openrouter-api-key=${OPENROUTER_API_KEY}" \
+            "xmind-api-key-openai=${XMIND_API_KEY_OPENAI}" \
+            "xmind-api-key-anthropic=${XMIND_API_KEY_ANTHROPIC}" \
+            "xmind-api-key-vertex-ai=${XMIND_API_KEY_VERTEX_AI}" \
            "ui-password=${UI_PASSWORD:-TaijiLiteLLM@2026!}"
    
    log_info "Secrets 更新完成"