$ cat /var/log/thoughts

AI Coding + Comparison

Posts matching all selected tags.

Filters2

Showing page 1 of 1 (3 total posts)

What Opus 5's Novel-Level Reasoning Benchmark Actually Means

Opus 5 novel reasoning rests on one number, 30.2% on ARC-AGI-3, and it scores action efficiency, not problems solved. Here are the four config lines to change.

Meta AnalysisAI CodingClaude CodeMLComparisonDeveloper Tools1 min read · Jul 25, 2026

How to Reduce AI Coding Tool Token Usage by 50%

Reduce AI coding tool token usage by half with system prompt trimming, targeted reads, and model routing. Save $100+/month on Claude Code, Cursor, and Copilot.

GuidesAI CodingClaude CodeDeveloper ToolsProductivityComparison1 min read · Apr 3, 2026

Claude Usage Limits Are a Bigger Problem Than Anthropic Admits

Claude usage limits burned $200 plans in 19 minutes. Zero published quotas, peak-hour rationing, and opacity are driving developers away from Anthropic.

OpinionAI CodingClaude CodeDeveloper ToolsTrend CommentaryComparison1 min read · Apr 1, 2026