🌱 Tool Showdowns · Head-to-head comparisons · cozy lesson
Choosing an LLM API
10 min · 1 min read · no scary math, promise
The one-line verdict
For your first app it barely matters: pick whoever has the clearest quickstart and a free tier today — OpenAI, Anthropic, or Gemini. What matters is keeping your code portable so switching costs an afternoon, not a rewrite.
Compare on these axes (check current pages — they move fast)
- Price per 1M input/output tokens — output tokens cost more everywhere.
- Rate limits on the free/tier-1 plan — prototypes die on 429s, not on quality.
- Structured output support — JSON mode keeps apps parseable.
- Context window you actually need — RAG with top-K=4 rarely needs the max.
Stay portable: one wrapper
def chat(messages, model="first-pick"):
# OpenAI-compatible shape; Anthropic/Gemini map onto it
return client.chat.completions.create(
model=model, messages=messages, temperature=0.2,
).choices[0].message.content
Evaluate all three on your own 10-question golden set (see Evals Basics), then commit. Re-run the set whenever you switch — 30 minutes that saves weeks of vibes-based regret.
Check your understanding
Correct answers earn XP (once each).
1. Best way to avoid lock-in?
2. What should drive the pick for a beginner?
My notes (saved in this browser)
Select text above → Save selection, or write your own. AlgoMaster-style notebook, local-first for MVP.
No notes yet. Your highlights will live here.
Finished reading? Seal it with a tick ✅
The checkbox in the explorer turns green too — same progress.