84% of our Anthropic API cost came from cache reads and writes.
Not input tokens, not output tokens.
Cache.
At Google, I honestly never thought much about the cost of AI. Joining an early stage startup completely changed that. Every dollar matters, so I became pretty obsessed with understanding where our token usage was actually going.
Most of our usage made sense. Our team averages around 70 to 100 PRs a day, so a large token bill is expected.
Then Fable was released and something felt very wrong. We were producing fewer PRs, but our cost was nearly 20x higher, and yes, I know what you’re thinking. Fable 5 is much more expensive than opus, but even after accounting for that, the math still didn’t work.
So I spent a week tracking everything claude code reported and comparing it against the Anthropic platform. The numbers mostly matched.
The surprising part was where the money was going. 84% of our usage came from cache reads and writes, split almost evenly, with cache reads slightly ahead.
In theory, caching should make things cheaper. In long running sessions, especially when claude code is acting as the main orchestrator, it became the biggest part of our bill.
Compaction helped, but only briefly. One request later, cache reads and writes would jump again. So I changed the workflow.
I now use Fable for planning, create the Linear ticket, then start a completely fresh session with Sonnet and let Codex handle the implementation. That simple change reduced our cost by around 50%.
Biggest takeaway:
If you’re paying for API usage, don’t just watch input and output tokens. Watch cache reads and writes very carefully.
Honestly, I suspect something similar happens on Claude subscriptions too. You just never see the bill.
For my first post, I’m sharing a letter @NVIDIA signed on why open models matter.
AI will transform every industry, power every company, and be built by every country.
Open models strengthen safety and cybersecurity, accelerate innovation and diffusion, and enable sovereignty.
The world needs both frontier closed models and frontier open models.
https://t.co/AUKzoQ5Ikb
Overlooked fact: every significant relationship in your life began with one moment of someone being slightly braver than the situation required. One person said the thing. Asked the question. Made the first move. Relationships don't start with mutual courage. They start with one person's excess courage.
I saw a quote that said "to procrastinate is to willingly endure the discomfort of anticipation, rather than the discomfort of action. Both are burdens, but only one leads to progress" and if that didn't light a fire in me.
New Anthropic research: Natural Language Autoencoders.
Models like Claude talk in words but think in numbers. The numbers—called activations—encode Claude’s thoughts, but not in a language we can read.
Here, we train Claude to translate its activations into human-readable text.
أنا متشخص GAD (قلق عام)، ومعرفش من العالم غير إني لما بفتح عيني من النوم أول حاجة بفكر فيها:
- ينهاااار أسود… هنعمل إيه في الموضوع ده؟
= تمام هظبطه، بس هو أنهي موضوع؟
- الموضوع المهم اللي أنت ناسيه دلوقتي.. يالهوااااي
= أنهي موضوع؟
- *مغص* .. *brain fog* .. *ضيق تنفس*…
Society begs men to open up, cry, and be emotionally vulnerable, but the silent, brutal truth nobody wants to admit is that a large percentage of women immediately lose romantic respect for a man the second he actually does. We call it "getting the ick." We essentially trick men into taking off their heavy armor just to judge them for the scars underneath, and then we have the audacity to wonder why they go right back to being stoic and completely emotionally closed off
رد فعل كوميدي من نسرين أمين لمّا عرفت إنها بتصور إعلان غسول 😂
اتفرج على برنامج #الكاميرا_الخفية حصريًا على #WATCHIT
من هنا https://t.co/6Dxp2hojEi
اشترك الآن بباقات جديدة تبدأ من ١٩.٩٩ج في الشهر ١٩٩.٩٩ج في السنة! https://t.co/vTMYn7hh6X
#رمضانك_عندنا#رمضان_2026#برامج_رمضان
Three years of freelancing taught me more about what I did not want than what I did. Decent clients, decent work, but no real identity to what I was building. Every project felt disconnected from the last one.
I'll be documenting what building this actually looks like. The decisions, the client work, the thinking behind it, and the things that don't go as planned. If any of that sounds relevant, follow along.
@tasmim_io
https://t.co/FrhglMnFGa
I came back to Cairo with a clearer head and started TASMIM, a digital systems studio. Premium websites and intelligent AI automations for agencies and consultants who are strong at what they do but whose digital setup hasn't kept up with their work.
Claude can now build interactive charts and diagrams, directly in the chat.
Available today in beta on all plans, including free.
Try it out: https://t.co/tHPAZRgQkn
TASMIM is a boutique digital systems studio. We build premium websites and intelligent AI automations for agencies and consultants, designed from the start to work together as one system.
Announcing Personal Computer.
Personal Computer is an always on, local merge with Perplexity Computer that works for you 24/7.
It's personal, secure, and works across your files, apps, and sessions through a continuously running Mac mini.