[AI] DeepSeek R1 和 Claude 3.5 Sonnet 的组合被认为是目前最强的模型组合，支持 OpenAI 流式输出，并可在 ChatBox 上运行

Summary

• ‣

• 项目支持本地运行和服务器运行，本地运行可与 Ollama 搭配，实现用本地的 DeepSeek R1 与 Claude 组合输出

git clone https://github.com/ErlichLiu/DeepClaude
cd DeepClaude

# see private
# change .env 

# add docker-compose.yml
cat > docker-compose.yml <<EOF
services:
  deepclaude:
    build:
      context: .
      dockerfile: Dockerfile
    ports:
      - "45678:8000"
    container_name: deepclaude
    env_file:
      - .env
    volumes:
      - ./app:/app/app  # 开发时方便代码热更新，生产部署可注释掉
    restart: unless-stopped
    #networks:
    #default:
    #name: npm_nginx_proxy_manager-network
    #external: true
EOF

# fix empty messages
git diff main
diff --git a/app/main.py b/app/main.py
index 64ad2f5..ff62e2f 100644
--- a/app/main.py
+++ b/app/main.py
@@ -81,6 +81,9 @@ async def chat_completions(request: Request):
             IS_ORIGIN_REASONING
         )

+        # 去空
+        messages = [msg for msg in messages if msg.get("content", "").strip() != ""]
+
         # 4. 返回流式响应
         return StreamingResponse(
             deep_claude.chat_completions_with_stream(
             
             
docker compose down; docker compose up -d; sleep 1s; docker compose logs -f

# if need to rebuild
docker compose down; docker compose up -d --build; sleep 1s; docker compose logs -f

Test Claude API (check host eligibility)

ANTHROPIC_API_KEY="sk-ant-api03-E9is5kJ"
curl https://api.anthropic.com/v1/messages \
     --header "x-api-key: $ANTHROPIC_API_KEY" \
     --header "anthropic-version: 2023-06-01" \
     --header "content-type: application/json" \
     --data \
'{
    "model": "claude-3-5-sonnet-20241022",
    "max_tokens": 1024,
    "messages": [
        {"role": "user", "content": "Hello, world"}
    ]
}'

Test Deepseek API (check host eligibility)

DEEPSEEK_API=
curl https://api.deepseek.com/chat/completions \
  -H "Content-Type: application/json" \
  -H "Authorization: Bearer $DEEPSEEK_API" \
  -d '{
        "model": "deepseek-chat",
        "messages": [
          {"role": "system", "content": "You are a helpful assistant."},
          {"role": "user", "content": "Hello!"}
        ],
        "stream": false
      }'