The fastest way to add AI to your app is to change one string.

#ai #api #llm #programming

Not "install this SDK." Not "create a new account." Not "read 40 pages of docs."
One string.

Here's the string:

# Before: locked to one provider
base_url="https://api.deepseek.com/v1"

# After: one endpoint, all models
base_url="https://aibridge-api.com/v1"

That's it. Your openai imports. Your chat.completions.create() calls.
Your response parsing. Your streaming logic. Nothing else changes.

import openai
client = openai.OpenAI(
    api_key="mb-xxx",
    base_url="https://aibridge-api.com/v1"
)

# deepseek-chat — $0.27/M, fastest
client.chat.completions.create(model="deepseek-chat", messages=messages)

# kimi-k3 — 1M context, always reasoning
client.chat.completions.create(model="kimi-k3", messages=messages)

# qwen-max — best multilingual
client.chat.completions.create(model="qwen-max", messages=messages)

Same library. Same signatures. Same response format.
Different model string. That's the entire "migration."

No new npm packages. No new environment variables.
No "let me just refactor this real quick" that turns into three hours.