How do I cut AI costs when using Claude or Codex all day?
Short answer
Use the top model for planning and a cheaper model for doing, avoid burning tokens on images and video, and default to the smaller model unless the task needs depth. My standing rule: Sonnet by default, Opus for research. And remember the $200 plans hide thousands of dollars of API-equivalent usage.
I jumped to the $200 a month Codex plan and used 80% of it in 36 hours. I need better AI cost discipline, and I say so openly. Here's what I've learned from my own bills and from builders I talk to.
Are the $200 a month plans worth it?
Usually, because they're subsidized. A co-founder and I compared notes: I cited roughly $7,000 to $8,000 of Claude usage and about $14,000 of ChatGPT usage, in API-equivalent terms, under flat monthly plans. Flat plans hide a lot of usage. You can test the coding tools on a $20 plan, though you'll get cut off quickly.
Which model should I use for what?
| Job | Model |
|---|---|
| Planning a build | Top model |
| Implementing the plan | Cheaper model |
| Everyday chats | Sonnet, by default |
| Newsletter research prompts | Opus |
The Sonnet default is a standing rule across my business because it gets three to five times more chats before you hit usage limits. My media buyer pushed back that Sonnet doesn't research as thoroughly, and I agreed, so research gets Opus.
What burns the most tokens?
Images and video. Those are the main token burners.
What do other builders do to save money?
These are their numbers, not mine:
- One peer runs GitHub Copilot at about $39 a month on an older Claude model at medium reasoning and says he never hit a token limit in two-day sessions. He also says structured, dashed-list prompts use fewer tokens.
- Another contact codes 10 to 12 hours a day on a $40 Copilot plan, using frontier models for planning only.
How else can I cut the bill?
- Move repeat work off expensive reasoning. When my AI-plan capacity ran dry in August 2026, the advice was to move deterministic execution, the same steps every time, away from expensive reasoning models. Back then I was spending more than $200 a month on AI.
- Stay under free tiers for tools. When I gave a research agent Exa and Firecrawl, I made sure it stayed under their free tiers: research tools for AI agents.
- Bundle the small subscriptions. Lenny's bundle gets you a pile of AI tools for roughly $20 or $30 a month. A year of the dictation tool alone is worth about $240.
- Split planning from building. That's the biggest lever: plan with one AI, build with another.
Answered by TJ Larkin. From partner calls, team calls and client calls in 2026. Last updated .