1. Use an API Gateway Instead of Direct Keys
Stop juggling multiple API keys. An API gateway like Crazyrouter gives you one key for 600+ models — GPT, Claude, Gemini, DeepSeek, all through OpenAI-compatible format.
# One key, any model
export OPENAI_API_KEY="sk-your-gateway-key"
export OPENAI_BASE_URL="https://crazyrouter.com/v1"
2. Match Model to Task Complexity
Don't use GPT-4o for everything. Use cheap models for simple tasks:
| Task | Model | Cost/1M tokens |
|---|---|---|
| Tab completion | gpt-4o-mini | $0.15 |
| Daily coding | deepseek-chat | $0.14 |
| Complex review | claude-sonnet-4 | $3.00 |
This alone can cut your AI bill by 90%.
3. Set Environment Variables Once
Most AI tools (Cursor, Cline, Aider, Continue) auto-detect OPENAI_API_KEY and OPENAI_BASE_URL. Set them in your shell profile and forget about per-tool configuration.
4. Stream Responses for Better UX
from openai import OpenAI
client = OpenAI(base_url="https://crazyrouter.com/v1")
stream = client.chat.completions.create(
model="deepseek-chat",
messages=[{"role": "user", "content": "Explain async/await"}],
stream=True
)
for chunk in stream:
if chunk.choices[0].delta.content:
print(chunk.choices[0].delta.content, end="")
5. Self-Host Your AI Gateway
For teams, self-hosting gives you full control. OpenClaw + Crazyrouter = private AI gateway in one command:
curl -fsSL https://raw.githubusercontent.com/xujfcn/crazyrouter-openclaw/main/install.sh | bash
Connect it to Telegram, Discord, or Slack for team-wide AI access.
What AI API tips do you use? Drop them in the comments 👇
- Crazyrouter — One API key for 600+ models
- OpenClaw — Open-source AI gateway
Top comments (0)