AIToday
Large Language ModelsAI Coding AssistantsZenn AI/MLPublished: Nov 17, 2026, 16:00 JST

Claude Code, Codex, SWE-2 each fail differently, $340/mo user finds

Send on LINE
Claude Code, Codex, SWE-2 each fail differently, $340/mo user finds

3 Key Points

  1. What happened

    A developer running Claude Code (Ultra, $220/month), Codex CLI (Pro, $100/month) and Devin's SWE-2 (Pro, $20/month) reports that Claude Code loses accuracy past 200K tokens and hits weekly limits, Codex drifts off-spec, and SWE-2 stalls on sub-agent rate limits.

  2. Why it matters

    Because each CLI breaks in its own way, the author treats them as three colleagues with different personalities — Claude Code for design and control, Codex for hands-on iteration, SWE-2 for unattended completion — rather than competing for the single 'smartest' tool.

Not sure about something? Ask the AI

Questions and answers are published on this page.

Summaries like this, every morning.

Context & Analysis

The author's setup is not a chain of tools passing work along; it is three separate flows matched to the stage of development. Claude Code is used as an orchestrator that stops at design, writing a spec within 200K–300K tokens to avoid the accuracy drop and weekly limit the author says kicks in when everything is pushed through it. Codex takes that spec for implementation, or handles lightweight mock work directly in the terminal with constant feedback, because left alone it tends to head off in its own direction. SWE-2 is reserved for self-contained, clearly bounded tasks and left running overnight, since the author finds it more autonomous than Claude Code and more faithful to instructions than Codex, though sub-agent sessions appear to count toward rate limits. The conclusion is not a ranking of intelligence but a division of labor among tools with different personalities.

FAQ
How much does the author spend on AI coding tools each month?
About $340 in total: Claude Code Ultra at $220/month, ChatGPT Pro for Codex CLI at $100/month, and Devin Pro for SWE-2 at $20/month.
What is the author's recommended workflow for Claude Code and Codex?
Use Claude Code to finish the design within 200K–300K tokens and output a Markdown spec with invariants, then hand that spec to Codex for implementation. This keeps Codex from drifting and saves Claude Code's context and weekly limit.
When does SWE-2 (Devin CLI) work best?
For heavy tasks with clear end conditions — adding test code, broad refactoring, or investigating and fixing hard-to-trace bugs — handed off whole and left running unattended while the PC stays on.
Send on LINE

AI news that matters for your work, delivered every morning.

Pick your industry and the AI tools you use, and get news related to your work every day.

Free · 30 seconds with Google · unsubscribe anytimeWhat is AIToday? →

Ask AI

Ask AI anything about this article. The AI reads this article, earlier AIToday articles, and Wikipedia, and cites its sources. Q&As are published on this page for other readers too.

Questions and answers are published on this page.

Related Articles

Next articleApple narrows macOS full disk access to selected folders in macOS 27.3 update