Hey hey, Useful AI Coding links to read/watch for this week of April 15, 2026.
If you want to get such list every Wednesday, subscribe via the form at the bottom.
Seems like last few weeks in AI coding community are not about new features or LLMs. Internet is buzzing about PRICES, USAGE LIMITS, and PERFORMANCE of LLMs.
Who cares about new Claude Code versions, when the core tool is unstable: a lot of downtime, random quality issues, and token usage horror stories.
Codex answered with their own new $100/month plan, but then people also started complaining about theit $20/mo plan being used much quicker. Same pattern like Anthropic?
Finally, GLM model subscription prices were silently raised by Zhipu to almost Anthropic level. Without any notice.
So, does it mean the days of subsidizing and cheap quality LLMs are over? The only way to find out is to wait and see.
Meanwhile this week's news and tools from myself and the community, enjoy!
For Premium Members
I Tried Superpowers Plugin vs Plan Mode in Claude Code / Codex
23-minute comparison video. I'm a big fan of Plan Mode in Claude Code, but Superpowers plugin gives extra layer on top. Is it worth it? I've tried it and compared to Codex plan mode, too.
From my YouTube channel
I Tried "grill-me" Skill for Plan Mode. Wow.
I've seen this skill getting viral on Twitter but never actually tried. It pleasantly surprised me.
I (Re-)Tested Opus 4.6 vs Opus 4.5: Is Quality Degradation Real?
The whole internet is shouting about Opus 4.6 being nerfed and degraded, so I made a small comparison experiment.
I Tried Superpowers Plugin: "Advanced" Plan Mode in Claude Code
A short free version of the video above. If you're a fan of deeper plan mode or spec-driven development, Superpowers may be a massive help.
AI Coding Community
OpenAI on X: "We’re introducing a new $100/month Pro tier."
x.com
We’re introducing a new $100/month Pro tier. This new tier offers 5x more Codex usage than Plus and is best for longer, high-effort Codex sessions.
You have made it completely unusable overnight. Please fix it. Thanks.
Ibo on X: "GLM prices: Chinese models used to be the cheap alternative"
x.com
Why are they priced like claude & codex now?
Kun Chen on X: "For folks who feel their Claude Code got nerfed"
x.com
Here's what I've set in my ~/.claude/settings.json to make my CC's behavior more stable. snippet and explanation in thread below.
Claude Code is completely down right now.
x.com
500 internal server errors. Not rate limits. The servers are crashing. Everyone is getting hit. Middle of a Monday. Peak hours. $200/month Max plan. 529 errors last week. 401 auth errors the week before. Now 500s. This is
Akshay on X: "A single CLAUDE.md file just hit 15K GitHub stars."
x.com
(derived from Karpathy's coding rules) Andrej Karpathy observed that LLMs make the same predictable mistakes when writing code: over-engineering, ignoring existing patterns, and adding dependencies you never asked for.
AI Benchmark Tool - Stupid Meter
aistupidlevel.info
The ultimate AI benchmarking tool for 2026. Compare AI models, test AI performance, and find the best AI for coding. Features real-time AI benchmark results, LLM performance tests, and comprehensive AI model leaderboard.
This Claude Skill Cuts Your Token Costs In HALF
youtube.com
The Caveman skill is a trending Claude Code prompt technique that strips filler words, hedging, and pleasantries from AI responses, cutting output tokens
That's it for this week, see you in the next newsletter issue!
Again, if you wanna get this to your inbox every Wednesday, subscribe via the form at the bottom.
