diff --git a/builds/mineru-vllm/docker-compose.yaml b/builds/mineru-vllm/docker-compose.yaml
deleted file mode 100644
index 102649c..0000000
--- a/builds/mineru-vllm/docker-compose.yaml
+++ /dev/null
@@ -1,109 +0,0 @@
-x-defaults: &defaults
-  restart: unless-stopped
-  logging:
-    driver: json-file
-    options:
-      max-size: 100m
-      max-file: "3"
-
-x-mineru-vllm: &mineru-vllm
-  <<: *defaults
-  image: ${GLOBAL_REGISTRY:-}alexsuntop/mineru:${MINERU_VERSION:-2.6.4}
-  build:
-    context: .
-    dockerfile: Dockerfile
-  environment:
-    TZ: ${TZ:-UTC}
-    MINERU_MODEL_SOURCE: local
-  ulimits:
-    memlock: -1
-    stack: 67108864
-  ipc: host
-  deploy:
-    resources:
-      limits:
-        cpus: '16.0'
-        memory: 32G
-      reservations:
-        cpus: '8.0'
-        memory: 16G
-        devices:
-          - driver: nvidia
-            device_ids: [ '0' ]
-            capabilities: [ gpu ]
-
-services:
-  mineru-vllm-server:
-    <<: *mineru-vllm
-    profiles: ["vllm-server"]
-    ports:
-      - ${MINERU_PORT_OVERRIDE_VLLM:-30000}:30000
-    entrypoint: mineru-vllm-server
-    command:
-      - --host 0.0.0.0
-      - --port 30000
-
-      # If using multiple GPUs, increase throughput using vllm's multi-GPU parallel mode.
-      # - --data-parallel-size 2
-      # If running on a single GPU and encountering VRAM shortage, reduce the KV cache size by this parameter,
-      # if VRAM issues persist, try lowering it further to `0.4` or below.
-      # - --gpu-memory-utilization 0.5
-
-    healthcheck:
-      test: ["CMD-SHELL", "curl -f http://localhost:30000/health || exit 1"]
-      interval: 30s
-      timeout: 10s
-      retries: 3
-      start_period: 60s
-
-
-  mineru-api:
-    <<: *mineru-vllm
-    profiles: ["api"]
-    ports:
-      - ${MINERU_PORT_OVERRIDE_API:-8000}:8000
-    entrypoint: mineru-api
-    command:
-      - --host 0.0.0.0
-      - --port 8000
-
-      # If using multiple GPUs, increase throughput using vllm's multi-GPU parallel mode.
-      # - --data-parallel-size 2
-      # If running on a single GPU and encountering VRAM shortage, reduce the KV cache size by this parameter,
-      # if VRAM issues persist, try lowering it further to `0.4` or below.
-      # - --gpu-memory-utilization 0.5
-    healthcheck:
-      test: ["CMD", "wget", "--no-verbose", "--tries=1", "--spider", "http://localhost:8000/health"]
-      interval: 30s
-      timeout: 10s
-      retries: 3
-      start_period: 60s
-
-  mineru-gradio:
-    <<: *mineru-vllm
-    profiles: ["gradio"]
-    ports:
-      - ${MINERU_PORT_OVERRIDE_GRADIO:-7860}:7860
-    entrypoint: mineru-gradio
-    command:
-      - --server-name 0.0.0.0
-      - --server-port 7860
-
-      # Enable the vllm engine for Gradio
-      - --enable-vllm-engine true
-      # If you want to disable the API, set this to false
-      # - --enable-api false
-      # If you want to limit the number of pages for conversion, set this to a specific number
-      # - --max-convert-pages 20
-
-      # If using multiple GPUs, increase throughput using vllm's multi-GPU parallel mode.
-      # - --data-parallel-size 2
-      # If running on a single GPU and encountering VRAM shortage, reduce the KV cache size by this parameter,
-      # if VRAM issues persist, try lowering it further to `0.4` or below.
-      # - --gpu-memory-utilization 0.5
-    healthcheck:
-      test: ["CMD", "wget", "--no-verbose", "--tries=1", "--spider", "http://localhost:7860/"]
-      interval: 30s
-      timeout: 10s
-      retries: 3
-      start_period: 60s
diff --git a/builds/mineru-vllm/.env.example b/builds/mineru/.env.example
similarity index 76%
rename from builds/mineru-vllm/.env.example
rename to builds/mineru/.env.example
index 47e5cb8..b121341 100644
--- a/builds/mineru-vllm/.env.example
+++ b/builds/mineru/.env.example
@@ -1,5 +1,5 @@
 # MinerU Docker image
-MINERU_DOCKER_IMAGE=alexsuntop/mineru:2.5.4
+MINERU_DOCKER_IMAGE=alexsuntop/mineru:2.6.5
 
 # Port configurations
 MINERU_PORT_OVERRIDE_VLLM=30000
diff --git a/builds/mineru-vllm/Dockerfile b/builds/mineru/Dockerfile
similarity index 100%
rename from builds/mineru-vllm/Dockerfile
rename to builds/mineru/Dockerfile
diff --git a/builds/mineru-vllm/README.md b/builds/mineru/README.md
similarity index 94%
rename from builds/mineru-vllm/README.md
rename to builds/mineru/README.md
index 61ea106..df5193f 100644
--- a/builds/mineru-vllm/README.md
+++ b/builds/mineru/README.md
@@ -39,7 +39,7 @@ mineru -p demo.pdf -o ./output -b vlm-http-client -u http://localhost:30000
 
 ## Configuration
 
-- `MINERU_VERSION`: The version for MinerU, default is `2.6.4`.
+- `MINERU_VERSION`: The version for MinerU, default is `2.6.5`.
 - `MINERU_PORT_OVERRIDE_VLLM`: The host port for the VLLM server, default is `30000`.
 - `MINERU_PORT_OVERRIDE_API`: The host port for the API service, default is `8000`.
 - `MINERU_PORT_OVERRIDE_GRADIO`: The host port for the Gradio WebUI, default is `7860`.
diff --git a/builds/mineru-vllm/README.zh.md b/builds/mineru/README.zh.md
similarity index 99%
rename from builds/mineru-vllm/README.zh.md
rename to builds/mineru/README.zh.md
index ea0c934..3b4ebb0 100644
--- a/builds/mineru-vllm/README.zh.md
+++ b/builds/mineru/README.zh.md
@@ -39,7 +39,7 @@ mineru -p demo.pdf -o ./output -b vlm-http-client -u http://localhost:30000
 
 ## 配置
 
-- `MINERU_VERSION`: MinerU 的 Docker 镜像版本，默认为 `2.6.4`。
+- `MINERU_VERSION`: MinerU 的 Docker 镜像版本，默认为 `2.6.5`。
 - `MINERU_PORT_OVERRIDE_VLLM`: VLLM 服务器的主机端口，默认为 `30000`。
 - `MINERU_PORT_OVERRIDE_API`: API 服务的主机端口，默认为 `8000`。
 - `MINERU_PORT_OVERRIDE_GRADIO`: Gradio WebUI 的主机端口，默认为 `7860`。
diff --git a/builds/mineru/docker-compose.yaml b/builds/mineru/docker-compose.yaml
new file mode 100644
index 0000000..63023fa
--- /dev/null
+++ b/builds/mineru/docker-compose.yaml
@@ -0,0 +1,139 @@
+x-defaults: &defaults
+  restart: unless-stopped
+  logging:
+    driver: json-file
+    options:
+      max-size: 100m
+      max-file: "3"
+
+x-mineru-vllm: &mineru-vllm
+  <<: *defaults
+  image: ${GLOBAL_REGISTRY:-}alexsuntop/mineru:${MINERU_VERSION:-2.6.5}
+  build:
+    context: .
+    dockerfile: Dockerfile
+  environment:
+    TZ: ${TZ:-UTC}
+    MINERU_MODEL_SOURCE: local
+  ulimits:
+    memlock: -1
+    stack: 67108864
+  ipc: host
+  deploy:
+    resources:
+      limits:
+        cpus: '16.0'
+        memory: 32G
+      reservations:
+        cpus: '8.0'
+        memory: 16G
+        devices:
+          - driver: nvidia
+            device_ids: [ '0' ]
+            capabilities: [ gpu ]
+
+services:
+  mineru-openai-server:
+    <<: *mineru-vllm
+    profiles: ["openai-server"]
+    ports:
+      - ${MINERU_PORT_OVERRIDE_VLLM:-30000}:30000
+    entrypoint: mineru-openai-server
+    command:
+      # ==================== Engine Selection ====================
+      # WARNING: Only ONE engine can be enabled at a time!
+      # Choose 'vllm' OR 'lmdeploy' (uncomment one line below)
+      - --engine vllm
+      # --engine lmdeploy
+
+      # ==================== vLLM Engine Parameters ====================
+      # Uncomment if using --engine vllm
+      - --host 0.0.0.0
+      - --port 30000
+      # Multi-GPU configuration (increase throughput)
+      # --data-parallel-size 2
+      # Single GPU memory optimization (reduce if VRAM insufficient)
+      # --gpu-memory-utilization 0.5  # Try 0.4 or lower if issues persist
+
+      # ==================== LMDeploy Engine Parameters ====================
+      # Uncomment if using --engine lmdeploy
+      # --server-name 0.0.0.0
+      # --server-port 30000
+      # Multi-GPU configuration (increase throughput)
+      # --dp 2
+      # Single GPU memory optimization (reduce if VRAM insufficient)
+      # --cache-max-entry-count 0.5  # Try 0.4 or lower if issues persist
+    healthcheck:
+      test: ["CMD-SHELL", "curl -f http://localhost:30000/health || exit 1"]
+      interval: 30s
+      timeout: 10s
+      retries: 3
+      start_period: 60s
+
+
+  mineru-api:
+    <<: *mineru-vllm
+    profiles: ["api"]
+    ports:
+      - ${MINERU_PORT_OVERRIDE_API:-8000}:8000
+    entrypoint: mineru-api
+    command:
+      # ==================== Server Configuration ====================
+      - --host 0.0.0.0
+      - --port 8000
+
+      # ==================== vLLM Engine Parameters ====================
+      # Multi-GPU configuration
+      # --data-parallel-size 2
+      # Single GPU memory optimization
+      # --gpu-memory-utilization 0.5  # Try 0.4 or lower if VRAM insufficient
+
+      # ==================== LMDeploy Engine Parameters ====================
+      # Multi-GPU configuration
+      # --dp 2
+      # Single GPU memory optimization
+      # --cache-max-entry-count 0.5  # Try 0.4 or lower if VRAM insufficient
+    healthcheck:
+      test: ["CMD", "wget", "--no-verbose", "--tries=1", "--spider", "http://localhost:8000/health"]
+      interval: 30s
+      timeout: 10s
+      retries: 3
+      start_period: 60s
+
+  mineru-gradio:
+    <<: *mineru-vllm
+    profiles: ["gradio"]
+    ports:
+      - ${MINERU_PORT_OVERRIDE_GRADIO:-7860}:7860
+    entrypoint: mineru-gradio
+    command:
+      # ==================== Gradio Server Configuration ====================
+      - --server-name 0.0.0.0
+      - --server-port 7860
+
+      # ==================== Gradio Feature Settings ====================
+      # --enable-api false  # Disable API endpoint
+      # --max-convert-pages 20  # Limit conversion page count
+
+      # ==================== Engine Selection ====================
+      # WARNING: Only ONE engine can be enabled at a time!
+
+      # Option 1: vLLM Engine (recommended for most users)
+      - --enable-vllm-engine true
+      # Multi-GPU configuration
+      # --data-parallel-size 2
+      # Single GPU memory optimization
+      # --gpu-memory-utilization 0.5  # Try 0.4 or lower if VRAM insufficient
+
+      # Option 2: LMDeploy Engine
+      # --enable-lmdeploy-engine true
+      # Multi-GPU configuration
+      # --dp 2
+      # Single GPU memory optimization
+      # --cache-max-entry-count 0.5  # Try 0.4 or lower if VRAM insufficient
+    healthcheck:
+      test: ["CMD", "wget", "--no-verbose", "--tries=1", "--spider", "http://localhost:7860/"]
+      interval: 30s
+      timeout: 10s
+      retries: 3
+      start_period: 60s
diff --git a/src/mineru-sglang/.env.example b/src/mineru-sglang/.env.example
deleted file mode 100644
index c8b3748..0000000
--- a/src/mineru-sglang/.env.example
+++ /dev/null
@@ -1,7 +0,0 @@
-# MinerU SGLang Docker image
-MINERU_DOCKER_IMAGE=alexsuntop/mineru-sglang:2.2.2
-
-# Port configurations
-MINERU_PORT_OVERRIDE_SGLANG=30000
-MINERU_PORT_OVERRIDE_API=8000
-MINERU_PORT_OVERRIDE_GRADIO=7860
diff --git a/src/mineru-sglang/README.md b/src/mineru-sglang/README.md
deleted file mode 100644
index 8695453..0000000
--- a/src/mineru-sglang/README.md
+++ /dev/null
@@ -1,45 +0,0 @@
-# MinerU SGLang
-
-[English](./README.md) | [中文](./README.zh.md)
-
-This service runs MinerU with the SGLang backend.
-
-## Start Services
-
-- **SGLang backend server**:
-
-  ```bash
-  docker compose --profile sglang-server up -d
-  ```
-
-- **Document parse API**:
-
-  ```bash
-  docker compose --profile api up -d
-  ```
-
-- **Gradio WebUI**:
-
-  ```bash
-  docker compose --profile gradio up -d
-  ```
-
-## Test SGLang backend
-
-```bash
-pip install mineru
-mineru -p demo.pdf -o ./output -b vlm-sglang-client -u http://localhost:30000
-```
-
-## Services
-
-- `mineru-sglang-server`: The SGLang backend server.
-- `mineru-api`: The document parsing API.
-- `mineru-gradio`: The Gradio WebUI.
-
-## Configuration
-
-- `MINERU_DOCKER_IMAGE`: The Docker image for MinerU SGLang, default is `alexsuntop/mineru-sglang:2.2.2`.
-- `MINERU_PORT_OVERRIDE_SGLANG`: The host port for the SGLang server, default is `30000`.
-- `MINERU_PORT_OVERRIDE_API`: The host port for the API service, default is `8000`.
-- `MINERU_PORT_OVERRIDE_GRADIO`: The host port for the Gradio WebUI, default is `7860`.
diff --git a/src/mineru-sglang/README.zh.md b/src/mineru-sglang/README.zh.md
deleted file mode 100644
index e7d3a97..0000000
--- a/src/mineru-sglang/README.zh.md
+++ /dev/null
@@ -1,45 +0,0 @@
-# MinerU SGLang
-
-[English](./README.md) | [中文](./README.zh.md)
-
-此服务使用 SGLang 后端运行 MinerU。
-
-## 启动服务
-
-- **SGLang 后端服务器**:
-
-  ```bash
-  docker compose --profile sglang-server up -d
-  ```
-
-- **文档解析 API**:
-
-  ```bash
-  docker compose --profile api up -d
-  ```
-
-- **Gradio WebUI**:
-
-  ```bash
-  docker compose --profile gradio up -d
-  ```
-
-## 测试 SGLang 后端
-
-```bash
-pip install mineru
-mineru -p demo.pdf -o ./output -b vlm-sglang-client -u http://localhost:30000
-```
-
-## 服务
-
-- `mineru-sglang-server`: SGLang 后端服务器。
-- `mineru-api`: 文档解析 API。
-- `mineru-gradio`: Gradio WebUI。
-
-## 配置
-
-- `MINERU_DOCKER_IMAGE`: MinerU SGLang 的 Docker 镜像，默认为 `alexsuntop/mineru-sglang:2.2.2`。
-- `MINERU_PORT_OVERRIDE_SGLANG`: SGLang 服务器的主机端口，默认为 `30000`。
-- `MINERU_PORT_OVERRIDE_API`: API 服务的主机端口，默认为 `8000`。
-- `MINERU_PORT_OVERRIDE_GRADIO`: Gradio WebUI 的主机端口，默认为 `7860`。
diff --git a/src/mineru-sglang/docker-compose.yaml b/src/mineru-sglang/docker-compose.yaml
deleted file mode 100644
index 74445ad..0000000
--- a/src/mineru-sglang/docker-compose.yaml
+++ /dev/null
@@ -1,105 +0,0 @@
-x-defaults: &defaults
-  restart: unless-stopped
-  logging:
-    driver: json-file
-    options:
-      max-size: 100m
-      max-file: "3"
-
-x-mineru-sglang: &mineru-sglang
-  <<: *defaults
-  image: ${MINERU_DOCKER_IMAGE:-alexsuntop/mineru-sglang:2.2.2}
-  environment:
-    TZ: ${TZ:-UTC}
-    MINERU_MODEL_SOURCE: local
-  ulimits:
-    memlock: -1
-    stack: 67108864
-  ipc: host
-  deploy:
-    resources:
-      limits:
-        cpus: ${MINERU_SGLANG_CPU_LIMIT:-8.0}
-        memory: ${MINERU_SGLANG_MEMORY_LIMIT:-4G}
-      reservations:
-        cpus: ${MINERU_SGLANG_CPU_RESERVATION:-1.0}
-        memory: ${MINERU_SGLANG_MEMORY_RESERVATION:-2G}
-        devices:
-          - driver: nvidia
-            device_ids: [ '0' ]
-            capabilities: [ gpu ]
-
-services:
-  mineru-sglang-server:
-    <<: *mineru-sglang
-    profiles: ["sglang-server"]
-    ports:
-      - ${MINERU_PORT_OVERRIDE_SGLANG:-30000}:30000
-    entrypoint: mineru-sglang-server
-    command:
-      - --host 0.0.0.0
-      - --port 30000
-
-      # If using multiple GPUs, increase throughput using vllm's multi-GPU parallel mode.
-      # - --data-parallel-size 2
-      # If running on a single GPU and encountering VRAM shortage, reduce the KV cache size by this parameter,
-      # if VRAM issues persist, try lowering it further to `0.4` or below.
-      # - --gpu-memory-utilization 0.5
-
-    healthcheck:
-      test: ["CMD-SHELL", "curl -f http://localhost:30000/health || exit 1"]
-      interval: 30s
-      timeout: 10s
-      retries: 3
-      start_period: 60s
-
-  mineru-api:
-    <<: *mineru-sglang
-    profiles: ["api"]
-    ports:
-      - ${MINERU_PORT_OVERRIDE_API:-8000}:8000
-    entrypoint: mineru-api
-    command:
-      - --host 0.0.0.0
-      - --port 8000
-
-      # If using multiple GPUs, increase throughput using vllm's multi-GPU parallel mode.
-      # - --data-parallel-size 2
-      # If running on a single GPU and encountering VRAM shortage, reduce the KV cache size by this parameter,
-      # if VRAM issues persist, try lowering it further to `0.4` or below.
-      # - --gpu-memory-utilization 0.5
-    healthcheck:
-      test: ["CMD", "wget", "--no-verbose", "--tries=1", "--spider", "http://localhost:8000/health"]
-      interval: 30s
-      timeout: 10s
-      retries: 3
-      start_period: 60s
-
-  mineru-gradio:
-    <<: *mineru-sglang
-    profiles: ["gradio"]
-    ports:
-      - ${MINERU_PORT_OVERRIDE_GRADIO:-7860}:7860
-    entrypoint: mineru-gradio
-    command:
-      - --server-name 0.0.0.0
-      - --server-port 7860
-
-      # Enable the vllm engine for Gradio
-      - --enable-vllm-engine true
-      # If you want to disable the API, set this to false
-      # - --enable-api false
-      # If you want to limit the number of pages for conversion, set this to a specific number
-      # - --max-convert-pages 20
-
-      # If using multiple GPUs, increase throughput using vllm's multi-GPU parallel mode.
-      # - --data-parallel-size 2
-      # If running on a single GPU and encountering VRAM shortage, reduce the KV cache size by this parameter,
-      # if VRAM issues persist, try lowering it further to `0.4` or below.
-      # - --gpu-memory-utilization 0.5
-    healthcheck:
-      test: ["CMD", "wget", "--no-verbose", "--tries=1", "--spider", "http://localhost:7860/"]
-      interval: 30s
-      timeout: 10s
-      retries: 3
-      start_period: 60s
diff --git a/src/open-webui-rust/.env.example b/src/open-webui-rust/.env.example
deleted file mode 100644
index e3e6e9c..0000000
--- a/src/open-webui-rust/.env.example
+++ /dev/null
@@ -1,133 +0,0 @@
-# Global Settings
-GLOBAL_REGISTRY=
-TZ=UTC
-
-# Version Tags
-POSTGRES_VERSION=16-alpine
-REDIS_VERSION=7-alpine
-RUST_BACKEND_VERSION=latest
-SANDBOX_RUNTIME_VERSION=latest
-SANDBOX_EXECUTOR_VERSION=latest
-FRONTEND_VERSION=latest
-
-# Port Overrides
-POSTGRES_PORT_OVERRIDE=5432
-REDIS_PORT_OVERRIDE=6379
-RUST_BACKEND_PORT_OVERRIDE=8080
-SANDBOX_EXECUTOR_PORT_OVERRIDE=8090
-FRONTEND_PORT_OVERRIDE=3000
-
-# PostgreSQL Configuration
-POSTGRES_DB=open_webui
-POSTGRES_USER=open_webui
-POSTGRES_PASSWORD=open_webui_password
-
-# PostgreSQL Resource Limits
-POSTGRES_CPU_LIMIT=1
-POSTGRES_MEMORY_LIMIT=1G
-POSTGRES_CPU_RESERVATION=0.25
-POSTGRES_MEMORY_RESERVATION=256M
-
-# Redis Resource Limits
-REDIS_CPU_LIMIT=0.5
-REDIS_MEMORY_LIMIT=512M
-REDIS_CPU_RESERVATION=0.1
-REDIS_MEMORY_RESERVATION=128M
-
-# Rust Backend Configuration
-ENV=prod
-WEBUI_SECRET_KEY=your-secret-key-min-32-chars-generate-with-uuidgen
-
-# Database Connection Pool
-DATABASE_POOL_SIZE=20
-DATABASE_POOL_MAX_OVERFLOW=10
-DATABASE_POOL_TIMEOUT=30
-DATABASE_POOL_RECYCLE=1800
-
-# Redis
-ENABLE_REDIS=true
-
-# Authentication
-JWT_EXPIRES_IN=30d
-ENABLE_SIGNUP=true
-ENABLE_LOGIN_FORM=true
-ENABLE_API_KEY=true
-DEFAULT_USER_ROLE=user
-SHOW_ADMIN_DETAILS=true
-WEBUI_URL=http://localhost:3000
-
-# CORS
-CORS_ALLOW_ORIGIN=*
-
-# WebSocket/Socket.IO
-ENABLE_SOCKETIO=true
-ENABLE_WEBSOCKET_SUPPORT=true
-WEBSOCKET_MANAGER=redis
-
-# Features
-ENABLE_OPENAI_API=true
-ENABLE_CHANNELS=true
-ENABLE_IMAGE_GENERATION=false
-ENABLE_CODE_EXECUTION=true
-ENABLE_CODE_INTERPRETER=true
-ENABLE_WEB_SEARCH=false
-ENABLE_ADMIN_CHAT_ACCESS=true
-ENABLE_ADMIN_EXPORT=true
-ENABLE_NOTES=true
-ENABLE_COMMUNITY_SHARING=true
-ENABLE_MESSAGE_RATING=true
-
-# Logging
-GLOBAL_LOG_LEVEL=info
-RUST_LOG=info
-
-# OpenAI (Optional)
-OPENAI_API_BASE_URL=
-OPENAI_API_KEY=
-
-# Rust Backend Resource Limits
-RUST_BACKEND_CPU_LIMIT=2
-RUST_BACKEND_MEMORY_LIMIT=2G
-RUST_BACKEND_CPU_RESERVATION=0.5
-RUST_BACKEND_MEMORY_RESERVATION=512M
-
-# Sandbox Executor Configuration
-SANDBOX_MAX_EXECUTION_TIME=60
-SANDBOX_MAX_MEMORY_MB=512
-SANDBOX_MAX_CPU_QUOTA=100000
-SANDBOX_MAX_DISK_MB=100
-SANDBOX_MAX_CONCURRENT_EXECUTIONS=10
-SANDBOX_RATE_LIMIT_PER_MINUTE=30
-SANDBOX_RATE_LIMIT_BURST=10
-SANDBOX_NETWORK_MODE=none
-SANDBOX_READ_ONLY_ROOT=false
-SANDBOX_DROP_ALL_CAPABILITIES=true
-SANDBOX_ENABLE_STREAMING=true
-SANDBOX_KEEP_CONTAINERS=false
-SANDBOX_ENABLE_PYTHON=true
-SANDBOX_ENABLE_JAVASCRIPT=true
-SANDBOX_ENABLE_SHELL=true
-SANDBOX_ENABLE_RUST=true
-SANDBOX_ENABLE_AUDIT_LOG=true
-
-# Sandbox Executor Resource Limits
-SANDBOX_EXECUTOR_CPU_LIMIT=2
-SANDBOX_EXECUTOR_MEMORY_LIMIT=2G
-SANDBOX_EXECUTOR_CPU_RESERVATION=0.25
-SANDBOX_EXECUTOR_MEMORY_RESERVATION=256M
-
-# Frontend Configuration
-RAG_EMBEDDING_MODEL=sentence-transformers/all-MiniLM-L6-v2
-RAG_RERANKING_MODEL=
-RAG_EMBEDDING_MODEL_AUTO_UPDATE=false
-WHISPER_MODEL=base
-ENABLE_RAG_WEB_SEARCH=false
-ENABLE_RAG_HYBRID_SEARCH=false
-SCARF_NO_ANALYTICS=true
-DO_NOT_TRACK=true
-
-# Frontend Resource Limits
-FRONTEND_CPU_LIMIT=1
-FRONTEND_MEMORY_LIMIT=1G
-FRONTEND_CPU_RESERVATION=0.25
-FRONTEND_MEMORY_RESERVATION=256M
diff --git a/src/open-webui-rust/README.md b/src/open-webui-rust/README.md
deleted file mode 100644
index 938a6d9..0000000
--- a/src/open-webui-rust/README.md
+++ /dev/null
@@ -1,345 +0,0 @@
-# Open WebUI Rust
-
-High-performance Rust implementation of Open WebUI with native async runtime and improved resource efficiency.
-
-## Overview
-
-Open WebUI Rust is a complete rewrite of the Open WebUI backend in Rust, offering:
-
-- **10-50x faster response times** for API endpoints
-- **70% lower memory usage** under load
-- **Native concurrency** with Tokio's async runtime
-- **Type safety** preventing entire classes of runtime errors
-- **Zero-copy streaming** for chat completions
-- **Production-ready** with comprehensive error handling
-
-This deployment includes:
-
-- **Rust Backend**: High-performance API server with WebSocket support
-- **PostgreSQL**: Primary database for data persistence
-- **Redis**: Caching and WebSocket session management
-- **Sandbox Executor**: Secure code execution environment with Docker isolation
-- **Frontend**: SvelteKit-based user interface with Nginx reverse proxy
-
-## Features
-
-- ✅ **Full API Compatibility** with Open WebUI Python backend
-- ✅ **Native WebSocket/Socket.IO** implementation in Rust
-- ✅ **Secure Code Execution** via isolated Docker containers
-- ✅ **Multi-language Support** (Python, JavaScript, Shell, Rust)
-- ✅ **RAG & Embeddings** with vector database integration
-- ✅ **Authentication & Authorization** with JWT tokens
-- ✅ **Rate Limiting & Security** built-in protection
-- ✅ **Comprehensive Logging** for debugging and monitoring
-
-## Prerequisites
-
-- Docker 20.10+
-- Docker Compose 2.0+
-- 4GB+ available memory
-- Access to Docker socket (for sandbox execution)
-
-## Quick Start
-
-### 1. Create `.env` file
-
-```bash
-cp .env.example .env
-```
-
-Edit `.env` and set required variables:
-
-```bash
-# IMPORTANT: Generate a secure secret key (min 32 characters)
-WEBUI_SECRET_KEY=$(uuidgen | tr '[:upper:]' '[:lower:]' | tr -d '-')
-
-# Optional: Configure OpenAI API
-OPENAI_API_KEY=sk-your-api-key
-OPENAI_API_BASE_URL=https://api.openai.com/v1
-```
-
-### 2. Start services
-
-```bash
-docker compose up -d
-```
-
-### 3. Access the application
-
-- **Frontend UI**: <http://localhost:3000>
-- **Rust Backend API**: <http://localhost:8080>
-- **Sandbox Executor**: <http://localhost:8090>
-
-### 4. Initial setup
-
-1. Open <http://localhost:3000>
-2. Create an admin account (first user becomes admin)
-3. Configure your AI models in Settings
-
-## Architecture
-
-```text
-┌─────────────┐
-│  Frontend   │ :3000
-│ (SvelteKit) │
-└──────┬──────┘
-       │
-       ↓
-┌─────────────┐     ┌──────────────┐
-│Rust Backend │────→│  PostgreSQL  │ :5432
-│   (API)     │     │  (Database)  │
-└──────┬──────┘     └──────────────┘
-       │
-       ├───────────→┌──────────────┐
-       │            │    Redis     │ :6379
-       │            │   (Cache)    │
-       │            └──────────────┘
-       │
-       └───────────→┌──────────────┐
-                    │   Sandbox    │ :8090
-                    │  Executor    │
-                    └──────────────┘
-```
-
-## Configuration
-
-### Core Environment Variables
-
-| Variable                     | Default               | Description                              |
-| ---------------------------- | --------------------- | ---------------------------------------- |
-| `WEBUI_SECRET_KEY`           | (required)            | Secret key for JWT tokens (min 32 chars) |
-| `POSTGRES_PASSWORD`          | `open_webui_password` | PostgreSQL password                      |
-| `FRONTEND_PORT_OVERRIDE`     | `3000`                | Frontend access port                     |
-| `RUST_BACKEND_PORT_OVERRIDE` | `8080`                | Backend API port                         |
-| `TZ`                         | `UTC`                 | Timezone setting                         |
-
-### Feature Flags
-
-| Variable                  | Default | Description                  |
-| ------------------------- | ------- | ---------------------------- |
-| `ENABLE_CODE_EXECUTION`   | `true`  | Enable secure code execution |
-| `ENABLE_CODE_INTERPRETER` | `true`  | Enable code interpreter      |
-| `ENABLE_IMAGE_GENERATION` | `false` | Enable image generation      |
-| `ENABLE_WEB_SEARCH`       | `false` | Enable web search            |
-| `ENABLE_SIGNUP`           | `true`  | Allow new user registration  |
-
-### Resource Limits
-
-Each service has configurable CPU and memory limits:
-
-```bash
-# Rust Backend
-RUST_BACKEND_CPU_LIMIT=2
-RUST_BACKEND_MEMORY_LIMIT=2G
-
-# PostgreSQL
-POSTGRES_CPU_LIMIT=1
-POSTGRES_MEMORY_LIMIT=1G
-
-# Sandbox Executor
-SANDBOX_EXECUTOR_CPU_LIMIT=2
-SANDBOX_EXECUTOR_MEMORY_LIMIT=2G
-```
-
-### Sandbox Security
-
-Configure sandbox execution limits:
-
-```bash
-SANDBOX_MAX_EXECUTION_TIME=60        # Max execution time in seconds
-SANDBOX_MAX_MEMORY_MB=512           # Max memory per execution
-SANDBOX_MAX_CONCURRENT_EXECUTIONS=10 # Max parallel executions
-SANDBOX_NETWORK_MODE=none           # Disable network access
-```
-
-## Usage Examples
-
-### Basic Chat
-
-```bash
-curl -X POST http://localhost:8080/api/chat/completions \
-  -H "Authorization: Bearer YOUR_API_KEY" \
-  -H "Content-Type: application/json" \
-  -d '{
-    "model": "gpt-3.5-turbo",
-    "messages": [{"role": "user", "content": "Hello!"}]
-  }'
-```
-
-### Code Execution
-
-```bash
-curl -X POST http://localhost:8090/api/v1/execute \
-  -H "Content-Type: application/json" \
-  -d '{
-    "language": "python",
-    "code": "print(\"Hello from sandbox!\")"
-  }'
-```
-
-## Monitoring
-
-### View Logs
-
-```bash
-# All services
-docker compose logs -f
-
-# Specific service
-docker compose logs -f rust-backend
-docker compose logs -f sandbox-executor
-```
-
-### Check Service Health
-
-```bash
-# Rust Backend
-curl http://localhost:8080/health
-
-# Sandbox Executor
-curl http://localhost:8090/api/v1/health
-
-# PostgreSQL
-docker compose exec postgres pg_isready
-```
-
-### Resource Usage
-
-```bash
-docker stats
-```
-
-## Maintenance
-
-### Backup Database
-
-```bash
-docker compose exec postgres pg_dump -U open_webui open_webui > backup.sql
-```
-
-### Restore Database
-
-```bash
-cat backup.sql | docker compose exec -T postgres psql -U open_webui open_webui
-```
-
-### Update Services
-
-```bash
-docker compose pull
-docker compose up -d
-```
-
-### Clean Reset
-
-```bash
-# Stop and remove all data
-docker compose down -v
-
-# Start fresh
-docker compose up -d
-```
-
-## Troubleshooting
-
-### Cannot connect to services
-
-Check if all containers are healthy:
-
-```bash
-docker compose ps
-```
-
-### JWT token errors
-
-Ensure `WEBUI_SECRET_KEY` is set and consistent across restarts.
-
-### Code execution fails
-
-1. Verify Docker socket is accessible
-2. Check sandbox executor logs: `docker compose logs sandbox-executor`
-3. Ensure runtime image is available: `docker images | grep sandbox-runtime`
-
-### High memory usage
-
-Adjust resource limits in `.env`:
-
-```bash
-RUST_BACKEND_MEMORY_LIMIT=1G
-SANDBOX_EXECUTOR_MEMORY_LIMIT=1G
-```
-
-## Security Considerations
-
-### Production Deployment
-
-1. **Change default passwords**
-
-   ```bash
-   POSTGRES_PASSWORD=$(openssl rand -base64 32)
-   WEBUI_SECRET_KEY=$(openssl rand -base64 48)
-   ```
-
-2. **Restrict network access**
-
-   ```bash
-   CORS_ALLOW_ORIGIN=https://yourdomain.com
-   ```
-
-3. **Disable signup after initial setup**
-
-   ```bash
-   ENABLE_SIGNUP=false
-   ```
-
-4. **Enable HTTPS** using a reverse proxy (nginx, Traefik, Caddy)
-
-5. **Limit sandbox resources**
-
-   ```bash
-   SANDBOX_MAX_EXECUTION_TIME=30
-   SANDBOX_MAX_MEMORY_MB=256
-   SANDBOX_NETWORK_MODE=none
-   ```
-
-### Docker Socket Access
-
-The sandbox executor requires access to the Docker socket (`/var/run/docker.sock`) for container isolation. This is a privileged operation and should be:
-
-- Used only in trusted environments
-- Protected with proper network isolation
-- Monitored for suspicious activity
-
-Consider using alternatives like:
-
-- Docker-in-Docker (DinD) for better isolation
-- Kubernetes with pod security policies
-- Dedicated sandbox service on separate nodes
-
-## Performance
-
-Typical resource usage:
-
-| Service          | CPU (idle) | Memory (idle) | CPU (load) | Memory (load) |
-| ---------------- | ---------- | ------------- | ---------- | ------------- |
-| Rust Backend     | ~1%        | 50MB          | 10-30%     | 200MB         |
-| PostgreSQL       | ~1%        | 50MB          | 5-15%      | 300MB         |
-| Redis            | <1%        | 10MB          | 2-5%       | 50MB          |
-| Sandbox Executor | <1%        | 30MB          | varies     | 512MB         |
-| Frontend         | <1%        | 50MB          | 5-10%      | 200MB         |
-
-## License
-
-This project follows the original Open WebUI license.
-
-## Related Projects
-
-- [Open WebUI](https://github.com/open-webui/open-webui) - Original Python implementation
-- [Open WebUI Rust](https://github.com/knoxchat/open-webui-rust) - Official Rust backend repository
-
-## Support
-
-For issues and questions:
-
-- [Open WebUI Rust Issues](https://github.com/knoxchat/open-webui-rust/issues)
-- [Open WebUI Documentation](https://docs.openwebui.com/)
diff --git a/src/open-webui-rust/README.zh.md b/src/open-webui-rust/README.zh.md
deleted file mode 100644
index 2111bfe..0000000
--- a/src/open-webui-rust/README.zh.md
+++ /dev/null
@@ -1,345 +0,0 @@
-# Open WebUI Rust
-
-基于 Rust 的高性能 Open WebUI 实现，具有原生异步运行时和优化的资源效率。
-
-## 概述
-
-Open WebUI Rust 是 Open WebUI 后端的完整 Rust 重写版本，提供：
-
-- **10-50 倍更快的响应时间**：API 端点性能显著提升
-- **70% 更低的内存使用**：负载下内存占用大幅降低
-- **原生并发**：基于 Tokio 的异步运行时
-- **类型安全**：防止整类运行时错误
-- **零拷贝流式传输**：聊天补全高效传输
-- **生产就绪**：全面的错误处理机制
-
-此部署包含：
-
-- **Rust 后端**：高性能 API 服务器，支持 WebSocket
-- **PostgreSQL**：数据持久化的主数据库
-- **Redis**：缓存和 WebSocket 会话管理
-- **Sandbox Executor**：安全的代码执行环境，具有 Docker 隔离
-- **Frontend**：基于 SvelteKit 的用户界面，配备 Nginx 反向代理
-
-## 特性
-
-- ✅ **完全兼容** Open WebUI Python 后端的 API
-- ✅ **原生 WebSocket/Socket.IO** 使用 Rust 实现
-- ✅ **安全的代码执行**：通过隔离的 Docker 容器
-- ✅ **多语言支持**：Python、JavaScript、Shell、Rust
-- ✅ **RAG 和嵌入**：集成向量数据库
-- ✅ **身份验证和授权**：JWT 令牌
-- ✅ **速率限制和安全**：内置保护机制
-- ✅ **全面的日志记录**：用于调试和监控
-
-## 前置要求
-
-- Docker 20.10+
-- Docker Compose 2.0+
-- 4GB+ 可用内存
-- 访问 Docker 套接字（用于沙箱执行）
-
-## 快速开始
-
-### 1. 创建 `.env` 文件
-
-```bash
-cp .env.example .env
-```
-
-编辑 `.env` 并设置必需的变量：
-
-```bash
-# 重要：生成一个安全的密钥（至少 32 个字符）
-WEBUI_SECRET_KEY=$(uuidgen | tr '[:upper:]' '[:lower:]' | tr -d '-')
-
-# 可选：配置 OpenAI API
-OPENAI_API_KEY=sk-your-api-key
-OPENAI_API_BASE_URL=https://api.openai.com/v1
-```
-
-### 2. 启动服务
-
-```bash
-docker compose up -d
-```
-
-### 3. 访问应用程序
-
-- **前端界面**：<http://localhost:3000>
-- **Rust 后端 API**：<http://localhost:8080>
-- **Sandbox Executor**：<http://localhost:8090>
-
-### 4. 初始设置
-
-1. 打开 <http://localhost:3000>
-2. 创建管理员账户（第一个用户成为管理员）
-3. 在设置中配置您的 AI 模型
-
-## 架构
-
-```text
-┌─────────────┐
-│  Frontend   │ :3000
-│ (SvelteKit) │
-└──────┬──────┘
-       │
-       ↓
-┌─────────────┐     ┌──────────────┐
-│Rust Backend │────→│  PostgreSQL  │ :5432
-│   (API)     │     │  (Database)  │
-└──────┬──────┘     └──────────────┘
-       │
-       ├───────────→┌──────────────┐
-       │            │    Redis     │ :6379
-       │            │   (Cache)    │
-       │            └──────────────┘
-       │
-       └───────────→┌──────────────┐
-                    │   Sandbox    │ :8090
-                    │  Executor    │
-                    └──────────────┘
-```
-
-## 配置
-
-### 核心环境变量
-
-| 变量                         | 默认值                | 描述                             |
-| ---------------------------- | --------------------- | -------------------------------- |
-| `WEBUI_SECRET_KEY`           | （必需）              | JWT 令牌的密钥（至少 32 个字符） |
-| `POSTGRES_PASSWORD`          | `open_webui_password` | PostgreSQL 密码                  |
-| `FRONTEND_PORT_OVERRIDE`     | `3000`                | 前端访问端口                     |
-| `RUST_BACKEND_PORT_OVERRIDE` | `8080`                | 后端 API 端口                    |
-| `TZ`                         | `UTC`                 | 时区设置                         |
-
-### 功能开关
-
-| 变量                      | 默认值  | 描述             |
-| ------------------------- | ------- | ---------------- |
-| `ENABLE_CODE_EXECUTION`   | `true`  | 启用安全代码执行 |
-| `ENABLE_CODE_INTERPRETER` | `true`  | 启用代码解释器   |
-| `ENABLE_IMAGE_GENERATION` | `false` | 启用图像生成     |
-| `ENABLE_WEB_SEARCH`       | `false` | 启用网络搜索     |
-| `ENABLE_SIGNUP`           | `true`  | 允许新用户注册   |
-
-### 资源限制
-
-每个服务都有可配置的 CPU 和内存限制：
-
-```bash
-# Rust 后端
-RUST_BACKEND_CPU_LIMIT=2
-RUST_BACKEND_MEMORY_LIMIT=2G
-
-# PostgreSQL
-POSTGRES_CPU_LIMIT=1
-POSTGRES_MEMORY_LIMIT=1G
-
-# Sandbox Executor
-SANDBOX_EXECUTOR_CPU_LIMIT=2
-SANDBOX_EXECUTOR_MEMORY_LIMIT=2G
-```
-
-### 沙箱安全
-
-配置沙箱执行限制：
-
-```bash
-SANDBOX_MAX_EXECUTION_TIME=60        # 最大执行时间（秒）
-SANDBOX_MAX_MEMORY_MB=512           # 每次执行的最大内存
-SANDBOX_MAX_CONCURRENT_EXECUTIONS=10 # 最大并行执行数
-SANDBOX_NETWORK_MODE=none           # 禁用网络访问
-```
-
-## 使用示例
-
-### 基础聊天
-
-```bash
-curl -X POST http://localhost:8080/api/chat/completions \
-  -H "Authorization: Bearer YOUR_API_KEY" \
-  -H "Content-Type: application/json" \
-  -d '{
-    "model": "gpt-3.5-turbo",
-    "messages": [{"role": "user", "content": "你好！"}]
-  }'
-```
-
-### 代码执行
-
-```bash
-curl -X POST http://localhost:8090/api/v1/execute \
-  -H "Content-Type: application/json" \
-  -d '{
-    "language": "python",
-    "code": "print(\"Hello from sandbox!\")"
-  }'
-```
-
-## 监控
-
-### 查看日志
-
-```bash
-# 所有服务
-docker compose logs -f
-
-# 特定服务
-docker compose logs -f rust-backend
-docker compose logs -f sandbox-executor
-```
-
-### 检查服务健康状态
-
-```bash
-# Rust 后端
-curl http://localhost:8080/health
-
-# Sandbox Executor
-curl http://localhost:8090/api/v1/health
-
-# PostgreSQL
-docker compose exec postgres pg_isready
-```
-
-### 资源使用情况
-
-```bash
-docker stats
-```
-
-## 维护
-
-### 备份数据库
-
-```bash
-docker compose exec postgres pg_dump -U open_webui open_webui > backup.sql
-```
-
-### 恢复数据库
-
-```bash
-cat backup.sql | docker compose exec -T postgres psql -U open_webui open_webui
-```
-
-### 更新服务
-
-```bash
-docker compose pull
-docker compose up -d
-```
-
-### 清理重置
-
-```bash
-# 停止并删除所有数据
-docker compose down -v
-
-# 全新启动
-docker compose up -d
-```
-
-## 故障排除
-
-### 无法连接到服务
-
-检查所有容器是否健康：
-
-```bash
-docker compose ps
-```
-
-### JWT 令牌错误
-
-确保 `WEBUI_SECRET_KEY` 已设置并在重启后保持一致。
-
-### 代码执行失败
-
-1. 验证 Docker 套接字是否可访问
-2. 检查 sandbox executor 日志：`docker compose logs sandbox-executor`
-3. 确保运行时镜像可用：`docker images | grep sandbox-runtime`
-
-### 内存使用过高
-
-在 `.env` 中调整资源限制：
-
-```bash
-RUST_BACKEND_MEMORY_LIMIT=1G
-SANDBOX_EXECUTOR_MEMORY_LIMIT=1G
-```
-
-## 安全注意事项
-
-### 生产部署
-
-1. **更改默认密码**
-
-   ```bash
-   POSTGRES_PASSWORD=$(openssl rand -base64 32)
-   WEBUI_SECRET_KEY=$(openssl rand -base64 48)
-   ```
-
-2. **限制网络访问**
-
-   ```bash
-   CORS_ALLOW_ORIGIN=https://yourdomain.com
-   ```
-
-3. **初始设置后禁用注册**
-
-   ```bash
-   ENABLE_SIGNUP=false
-   ```
-
-4. **使用反向代理启用 HTTPS**（nginx、Traefik、Caddy）
-
-5. **限制沙箱资源**
-
-   ```bash
-   SANDBOX_MAX_EXECUTION_TIME=30
-   SANDBOX_MAX_MEMORY_MB=256
-   SANDBOX_NETWORK_MODE=none
-   ```
-
-### Docker 套接字访问
-
-Sandbox Executor 需要访问 Docker 套接字（`/var/run/docker.sock`）以实现容器隔离。这是一个特权操作，应该：
-
-- 仅在受信任的环境中使用
-- 通过适当的网络隔离进行保护
-- 监控可疑活动
-
-考虑使用以下替代方案：
-
-- Docker-in-Docker（DinD）以获得更好的隔离
-- 带有 Pod 安全策略的 Kubernetes
-- 在单独节点上的专用沙箱服务
-
-## 性能
-
-典型资源使用情况：
-
-| 服务             | CPU（空闲） | 内存（空闲） | CPU（负载） | 内存（负载） |
-| ---------------- | ----------- | ------------ | ----------- | ------------ |
-| Rust Backend     | ~1%         | 50MB         | 10-30%      | 200MB        |
-| PostgreSQL       | ~1%         | 50MB         | 5-15%       | 300MB        |
-| Redis            | <1%         | 10MB         | 2-5%        | 50MB         |
-| Sandbox Executor | <1%         | 30MB         | 变化        | 512MB        |
-| Frontend         | <1%         | 50MB         | 5-10%       | 200MB        |
-
-## 许可证
-
-本项目遵循原 Open WebUI 的许可证。
-
-## 相关项目
-
-- [Open WebUI](https://github.com/open-webui/open-webui) - 原始 Python 实现
-- [Open WebUI Rust](https://github.com/knoxchat/open-webui-rust) - 官方 Rust 后端仓库
-
-## 支持
-
-问题和疑问：
-
-- [Open WebUI Rust Issues](https://github.com/knoxchat/open-webui-rust/issues)
-- [Open WebUI 文档](https://docs.openwebui.com/)
diff --git a/src/open-webui-rust/docker-compose.yaml b/src/open-webui-rust/docker-compose.yaml
deleted file mode 100644
index 80d892a..0000000
--- a/src/open-webui-rust/docker-compose.yaml
+++ /dev/null
@@ -1,310 +0,0 @@
-x-defaults: &defaults
-  restart: unless-stopped
-  logging:
-    driver: json-file
-    options:
-      max-size: 100m
-      max-file: "3"
-
-services:
-  # PostgreSQL Database
-  postgres:
-    <<: *defaults
-    image: ${GLOBAL_REGISTRY:-}postgres:${POSTGRES_VERSION:-16-alpine}
-    environment:
-      POSTGRES_DB: ${POSTGRES_DB:-open_webui}
-      POSTGRES_USER: ${POSTGRES_USER:-open_webui}
-      POSTGRES_PASSWORD: ${POSTGRES_PASSWORD:-open_webui_password}
-      POSTGRES_INITDB_ARGS: "-E UTF8"
-      TZ: ${TZ:-UTC}
-    volumes:
-      - postgres_data:/var/lib/postgresql/data
-    ports:
-      - "${POSTGRES_PORT_OVERRIDE:-5432}:5432"
-    healthcheck:
-      test: ["CMD-SHELL", "pg_isready -U ${POSTGRES_USER:-open_webui}"]
-      interval: 10s
-      timeout: 5s
-      retries: 5
-      start_period: 10s
-    deploy:
-      resources:
-        limits:
-          cpus: ${POSTGRES_CPU_LIMIT:-1}
-          memory: ${POSTGRES_MEMORY_LIMIT:-1G}
-        reservations:
-          cpus: ${POSTGRES_CPU_RESERVATION:-0.25}
-          memory: ${POSTGRES_MEMORY_RESERVATION:-256M}
-    networks:
-      - open-webui-network
-
-  # Redis for caching and WebSocket management
-  redis:
-    <<: *defaults
-    image: ${GLOBAL_REGISTRY:-}redis:${REDIS_VERSION:-7-alpine}
-    command: redis-server --appendonly yes
-    environment:
-      TZ: ${TZ:-UTC}
-    volumes:
-      - redis_data:/data
-    ports:
-      - "${REDIS_PORT_OVERRIDE:-6379}:6379"
-    healthcheck:
-      test: ["CMD", "redis-cli", "ping"]
-      interval: 10s
-      timeout: 5s
-      retries: 5
-      start_period: 10s
-    deploy:
-      resources:
-        limits:
-          cpus: ${REDIS_CPU_LIMIT:-0.5}
-          memory: ${REDIS_MEMORY_LIMIT:-512M}
-        reservations:
-          cpus: ${REDIS_CPU_RESERVATION:-0.1}
-          memory: ${REDIS_MEMORY_RESERVATION:-128M}
-    networks:
-      - open-webui-network
-
-  # Rust Backend
-  rust-backend:
-    <<: *defaults
-    image: ${GLOBAL_REGISTRY:-}public.ecr.aws/o3p7x2f5/knoxchat/open-webui-rust-backend:${RUST_BACKEND_VERSION:-latest}
-    environment:
-      # Server
-      HOST: 0.0.0.0
-      PORT: 8080
-      ENV: ${ENV:-prod}
-      WEBUI_SECRET_KEY: ${WEBUI_SECRET_KEY}
-      TZ: ${TZ:-UTC}
-
-      # Database
-      DATABASE_URL: postgresql://${POSTGRES_USER:-open_webui}:${POSTGRES_PASSWORD:-open_webui_password}@postgres:5432/${POSTGRES_DB:-open_webui}
-      DATABASE_POOL_SIZE: ${DATABASE_POOL_SIZE:-20}
-      DATABASE_POOL_MAX_OVERFLOW: ${DATABASE_POOL_MAX_OVERFLOW:-10}
-      DATABASE_POOL_TIMEOUT: ${DATABASE_POOL_TIMEOUT:-30}
-      DATABASE_POOL_RECYCLE: ${DATABASE_POOL_RECYCLE:-1800}
-
-      # Redis
-      ENABLE_REDIS: ${ENABLE_REDIS:-true}
-      REDIS_URL: redis://redis:6379
-
-      # Authentication
-      JWT_EXPIRES_IN: ${JWT_EXPIRES_IN:-30d}
-      ENABLE_SIGNUP: ${ENABLE_SIGNUP:-true}
-      ENABLE_LOGIN_FORM: ${ENABLE_LOGIN_FORM:-true}
-      ENABLE_API_KEY: ${ENABLE_API_KEY:-true}
-      DEFAULT_USER_ROLE: ${DEFAULT_USER_ROLE:-user}
-      SHOW_ADMIN_DETAILS: ${SHOW_ADMIN_DETAILS:-true}
-      WEBUI_URL: ${WEBUI_URL:-http://localhost:3000}
-
-      # CORS
-      CORS_ALLOW_ORIGIN: ${CORS_ALLOW_ORIGIN:-*}
-
-      # WebSocket/Socket.IO (Native Rust Implementation)
-      ENABLE_SOCKETIO: ${ENABLE_SOCKETIO:-true}
-      ENABLE_WEBSOCKET_SUPPORT: ${ENABLE_WEBSOCKET_SUPPORT:-true}
-      WEBSOCKET_MANAGER: ${WEBSOCKET_MANAGER:-redis}
-      WEBSOCKET_REDIS_URL: redis://redis:6379
-
-      # Features
-      ENABLE_OPENAI_API: ${ENABLE_OPENAI_API:-true}
-      ENABLE_CHANNELS: ${ENABLE_CHANNELS:-true}
-      ENABLE_IMAGE_GENERATION: ${ENABLE_IMAGE_GENERATION:-false}
-      ENABLE_CODE_EXECUTION: ${ENABLE_CODE_EXECUTION:-true}
-      CODE_EXECUTION_SANDBOX_URL: http://sandbox-executor:8090
-      ENABLE_CODE_INTERPRETER: ${ENABLE_CODE_INTERPRETER:-true}
-      CODE_INTERPRETER_SANDBOX_URL: http://sandbox-executor:8090
-      ENABLE_WEB_SEARCH: ${ENABLE_WEB_SEARCH:-false}
-      ENABLE_ADMIN_CHAT_ACCESS: ${ENABLE_ADMIN_CHAT_ACCESS:-true}
-      ENABLE_ADMIN_EXPORT: ${ENABLE_ADMIN_EXPORT:-true}
-      ENABLE_NOTES: ${ENABLE_NOTES:-true}
-      ENABLE_COMMUNITY_SHARING: ${ENABLE_COMMUNITY_SHARING:-true}
-      ENABLE_MESSAGE_RATING: ${ENABLE_MESSAGE_RATING:-true}
-
-      # Storage
-      UPLOAD_DIR: /app/data/uploads
-      CACHE_DIR: /app/data/cache
-      STATIC_DIR: /app/static
-
-      # Logging
-      GLOBAL_LOG_LEVEL: ${GLOBAL_LOG_LEVEL:-info}
-      RUST_LOG: ${RUST_LOG:-info}
-
-      # OpenAI
-      OPENAI_API_BASE_URL: ${OPENAI_API_BASE_URL:-}
-      OPENAI_API_KEY: ${OPENAI_API_KEY:-}
-
-    volumes:
-      - rust_backend_data:/app/data
-    ports:
-      - "${RUST_BACKEND_PORT_OVERRIDE:-8080}:8080"
-    depends_on:
-      postgres:
-        condition: service_healthy
-      redis:
-        condition: service_healthy
-      sandbox-executor:
-        condition: service_healthy
-    healthcheck:
-      test: ["CMD-SHELL", "curl -f http://localhost:8080/health || exit 1"]
-      interval: 30s
-      timeout: 10s
-      retries: 3
-      start_period: 40s
-    deploy:
-      resources:
-        limits:
-          cpus: ${RUST_BACKEND_CPU_LIMIT:-2}
-          memory: ${RUST_BACKEND_MEMORY_LIMIT:-2G}
-        reservations:
-          cpus: ${RUST_BACKEND_CPU_RESERVATION:-0.5}
-          memory: ${RUST_BACKEND_MEMORY_RESERVATION:-512M}
-    networks:
-      - open-webui-network
-
-  # Sandbox Runtime
-  sandbox-runtime:
-    <<: *defaults
-    image: ${GLOBAL_REGISTRY:-}public.ecr.aws/o3p7x2f5/knoxchat/sandbox-runtime:${SANDBOX_RUNTIME_VERSION:-latest}
-    command: /bin/true
-    environment:
-      TZ: ${TZ:-UTC}
-    networks:
-      - open-webui-network
-
-  # Sandbox Executor Service
-  sandbox-executor:
-    <<: *defaults
-    image: ${GLOBAL_REGISTRY:-}public.ecr.aws/o3p7x2f5/knoxchat/sandbox-executor:${SANDBOX_EXECUTOR_VERSION:-latest}
-    user: root
-    ports:
-      - "${SANDBOX_EXECUTOR_PORT_OVERRIDE:-8090}:8090"
-    volumes:
-      - /var/run/docker.sock:/var/run/docker.sock
-      - sandbox_logs:/var/log/sandbox-executor
-    environment:
-      SANDBOX_HOST: 0.0.0.0
-      SANDBOX_PORT: 8090
-      DOCKER_HOST: unix:///var/run/docker.sock
-      MAX_EXECUTION_TIME: ${SANDBOX_MAX_EXECUTION_TIME:-60}
-      MAX_MEMORY_MB: ${SANDBOX_MAX_MEMORY_MB:-512}
-      MAX_CPU_QUOTA: ${SANDBOX_MAX_CPU_QUOTA:-100000}
-      MAX_DISK_MB: ${SANDBOX_MAX_DISK_MB:-100}
-      MAX_CONCURRENT_EXECUTIONS: ${SANDBOX_MAX_CONCURRENT_EXECUTIONS:-10}
-      RATE_LIMIT_PER_MINUTE: ${SANDBOX_RATE_LIMIT_PER_MINUTE:-30}
-      RATE_LIMIT_BURST: ${SANDBOX_RATE_LIMIT_BURST:-10}
-      CONTAINER_IMAGE: public.ecr.aws/o3p7x2f5/knoxchat/sandbox-runtime:latest
-      NETWORK_MODE: ${SANDBOX_NETWORK_MODE:-none}
-      READ_ONLY_ROOT: ${SANDBOX_READ_ONLY_ROOT:-false}
-      DROP_ALL_CAPABILITIES: ${SANDBOX_DROP_ALL_CAPABILITIES:-true}
-      ENABLE_STREAMING: ${SANDBOX_ENABLE_STREAMING:-true}
-      KEEP_CONTAINERS: ${SANDBOX_KEEP_CONTAINERS:-false}
-      ENABLE_PYTHON: ${SANDBOX_ENABLE_PYTHON:-true}
-      ENABLE_JAVASCRIPT: ${SANDBOX_ENABLE_JAVASCRIPT:-true}
-      ENABLE_SHELL: ${SANDBOX_ENABLE_SHELL:-true}
-      ENABLE_RUST: ${SANDBOX_ENABLE_RUST:-true}
-      ENABLE_AUDIT_LOG: ${SANDBOX_ENABLE_AUDIT_LOG:-true}
-      AUDIT_LOG_PATH: /var/log/sandbox-executor/audit.log
-      TZ: ${TZ:-UTC}
-    depends_on:
-      - sandbox-runtime
-    healthcheck:
-      test: ["CMD", "curl", "-f", "http://localhost:8090/api/v1/health"]
-      interval: 30s
-      timeout: 10s
-      retries: 3
-      start_period: 10s
-    deploy:
-      resources:
-        limits:
-          cpus: ${SANDBOX_EXECUTOR_CPU_LIMIT:-2}
-          memory: ${SANDBOX_EXECUTOR_MEMORY_LIMIT:-2G}
-        reservations:
-          cpus: ${SANDBOX_EXECUTOR_CPU_RESERVATION:-0.25}
-          memory: ${SANDBOX_EXECUTOR_MEMORY_RESERVATION:-256M}
-    networks:
-      - open-webui-network
-
-  # Frontend (SvelteKit)
-  frontend:
-    <<: *defaults
-    image: ${GLOBAL_REGISTRY:-}public.ecr.aws/o3p7x2f5/knoxchat/open-webui-frontend:${FRONTEND_VERSION:-latest}
-    environment:
-      # Backend URLs (Socket.IO now served by Rust backend)
-      BACKEND_URL: http://rust-backend:8080
-      SANDBOX_EXECUTOR_URL: http://sandbox-executor:8090
-
-      # Server
-      ENV: ${ENV:-prod}
-      PORT: 8080
-      WEBUI_SECRET_KEY: ${WEBUI_SECRET_KEY}
-      TZ: ${TZ:-UTC}
-
-      # Database
-      DATABASE_URL: postgresql://${POSTGRES_USER:-open_webui}:${POSTGRES_PASSWORD:-open_webui_password}@postgres:5432/${POSTGRES_DB:-open_webui}
-
-      # OpenAI
-      OPENAI_API_BASE_URL: ${OPENAI_API_BASE_URL:-}
-      OPENAI_API_KEY: ${OPENAI_API_KEY:-}
-
-      # RAG & Embeddings
-      RAG_EMBEDDING_MODEL: ${RAG_EMBEDDING_MODEL:-sentence-transformers/all-MiniLM-L6-v2}
-      RAG_RERANKING_MODEL: ${RAG_RERANKING_MODEL:-}
-      RAG_EMBEDDING_MODEL_AUTO_UPDATE: ${RAG_EMBEDDING_MODEL_AUTO_UPDATE:-false}
-
-      # Whisper
-      WHISPER_MODEL: ${WHISPER_MODEL:-base}
-
-      # Redis
-      REDIS_URL: redis://redis:6379
-
-      # Features
-      ENABLE_RAG_WEB_SEARCH: ${ENABLE_RAG_WEB_SEARCH:-false}
-      ENABLE_RAG_HYBRID_SEARCH: ${ENABLE_RAG_HYBRID_SEARCH:-false}
-      ENABLE_IMAGE_GENERATION: ${ENABLE_IMAGE_GENERATION:-false}
-
-      # Analytics
-      SCARF_NO_ANALYTICS: ${SCARF_NO_ANALYTICS:-true}
-      DO_NOT_TRACK: ${DO_NOT_TRACK:-true}
-
-    volumes:
-      - frontend_data:/app/backend/data
-    ports:
-      - "${FRONTEND_PORT_OVERRIDE:-3000}:8080"
-    depends_on:
-      rust-backend:
-        condition: service_healthy
-      postgres:
-        condition: service_healthy
-      redis:
-        condition: service_healthy
-      sandbox-executor:
-        condition: service_healthy
-    healthcheck:
-      test: ["CMD-SHELL", "curl -f http://localhost:8080/health || exit 1"]
-      interval: 30s
-      timeout: 10s
-      retries: 3
-      start_period: 60s
-    deploy:
-      resources:
-        limits:
-          cpus: ${FRONTEND_CPU_LIMIT:-1}
-          memory: ${FRONTEND_MEMORY_LIMIT:-1G}
-        reservations:
-          cpus: ${FRONTEND_CPU_RESERVATION:-0.25}
-          memory: ${FRONTEND_MEMORY_RESERVATION:-256M}
-    networks:
-      - open-webui-network
-
-volumes:
-  postgres_data:
-  redis_data:
-  rust_backend_data:
-  sandbox_logs:
-  frontend_data:
-
-networks:
-  open-webui-network:
-    driver: bridge