I replaced ChatGPT with a $2/month API — here's my exact curl command

#claude #ai #programming #api

I replaced ChatGPT with a $2/month API — here's my exact curl command

I've been using Claude AI for my development workflow for about a year. The official Claude API costs serious money at scale, and ChatGPT Plus is $20/month — which adds up fast.

A few months ago I switched to SimplyLouie — a proxy API for Claude that costs $2/month flat. Here's the exact curl command I use every day:

curl https://simplylouie.com/api/chat \
  -H "Content-Type: application/json" \
  -H "X-API-Key: your-api-key-here" \
  -d '{
    "model": "claude-opus-4-5",
    "max_tokens": 1024,
    "messages": [
      {"role": "user", "content": "Explain this function and suggest improvements"}
    ]
  }'

Response looks like standard Anthropic API:

{
  "id": "msg_abc123",
  "type": "message",
  "role": "assistant",
  "content": [
    {
      "type": "text",
      "text": "Here's what this function does..."
    }
  ],
  "model": "claude-opus-4-5",
  "stop_reason": "end_turn"
}

Why I switched

The official Anthropic API: Pay per token. A heavy coding session (debugging a gnarly production incident, reviewing a large PR) can cost $5-15 in a single afternoon.

ChatGPT Plus: $20/month. Works fine, but I'm a Claude person — the reasoning quality on complex code is noticeably better for my use cases.

SimplyLouie: $2/month flat. Same Claude API. No token counting anxiety.

For me, the math was obvious.

Setting it up with Claude Code

If you use Claude Code (the CLI tool), you can point it at any OpenAI-compatible endpoint:

export ANTHROPIC_BASE_URL=https://simplylouie.com
export ANTHROPIC_API_KEY=your-api-key-here
claude

That's it. Claude Code works identically — same commands, same workflow, same output quality.

My daily workflow

Here's a real example. I'm doing a code review on a 400-line PR:

# Get the diff
git diff main...feature-branch > /tmp/pr-diff.txt

# Send to Claude via SimplyLouie
curl https://simplylouie.com/api/chat \
  -H "Content-Type: application/json" \
  -H "X-API-Key: $SIMPLYLOUIE_KEY" \
  -d "{
    \"model\": \"claude-opus-4-5\",
    \"max_tokens\": 4096,
    \"messages\": [{
      \"role\": \"user\",
      \"content\": \"Review this PR diff for bugs, security issues, and code quality:\\n\\n$(cat /tmp/pr-diff.txt)\"
    }]
  }"

I run this 10-15 times a day. On the official API, that would be expensive. On SimplyLouie, it's covered by my $2/month.

Python integration

For scripts, I use the Anthropic Python SDK pointed at the SimplyLouie endpoint:

import anthropic

client = anthropic.Anthropic(
    api_key="your-simplylouie-key",
    base_url="https://simplylouie.com"
)

message = client.messages.create(
    model="claude-opus-4-5",
    max_tokens=1024,
    messages=[
        {"role": "user", "content": "Refactor this function to be more readable"}
    ]
)

print(message.content[0].text)

What I use it for

PR reviews — paste the diff, get structured feedback
Debugging — paste error logs + stack traces, ask what's wrong
Writing commit messages — git diff --staged | claude-summarize
Documentation — paste a function, get JSDoc/docstring output
Code explanation — onboarding to unfamiliar codebases

All of these are heavy-token operations. The flat $2/month pricing means I don't think about cost — I just use it.

The catch?

Honestly, I haven't found one. The API is responsive, the model quality is the same (it's proxying to the same Anthropic backend), and the price is 10x cheaper than ChatGPT Plus.

The only thing to know: it's a proxy service, so you're trusting SimplyLouie to stay up and keep the lights on. For personal dev use, that's a fine tradeoff. For production systems, I'd use the official API.