DoomPosting
8.14K subscribers
90.8K photos
29.6K videos
6 files
90.6K links
Degens Deteriorating
Download Telegram
This media is not supported in your browser
VIEW IN TELEGRAM
Marines who’ve been stuck on a boat for a month realizing they don’t get to kill anyone

πŸ„³πŸ„ΎπŸ„ΎπŸ„ΌπŸ„ΏπŸ€–πŸ…‚πŸ…ƒπŸ„ΈπŸ„½πŸ„Ά
😁5🫑4
This media is not supported in your browser
VIEW IN TELEGRAM
>ceasefire announced

>Iran launches missiles

>oil prices unpredictable

>GOP pushing amnesty

Me: *heading back to racist Japanese 𝕏*

πŸ„³πŸ„ΎπŸ„ΎπŸ„ΌπŸ„ΏπŸ€–πŸ…‚πŸ…ƒπŸ„ΈπŸ„½πŸ„Ά
πŸ”₯5⚑4
Claude Code doesn't show you how many tokens you're using for subscriptions. No breakdown by model. No breakdown by project. Just a progress bar that says "63% used."

So I built a local dashboard that reads the files Claude Code already writes to your machine.

Turns out every session, every turn, every token is logged to ~/.claude/projects/ in JSONL files. Input tokens, output tokens, cache reads, cache creation, model name, timestamp.

It's all there. You just can't see it.

My numbers over the last 30 days: 440 sessions. 18,000 turns. $1,588 in API-equivalent costs. On one day, the cache spiked to 700M tokens - visible cache bug, two days in a row.

The dashboard scans those local files, builds a SQLite database, and serves charts on localhost:8080. Filter by model (Opus, Sonnet, Haiku). Filter by time range (7d, 30d, 90d, all time). Cost estimates based on current Anthropic API pricing.

Works retroactively. First run processes your entire Claude Code history.

Install:

git clone

πŸ„³πŸ„ΎπŸ„ΎπŸ„ΌπŸ„ΏπŸ€–πŸ…‚πŸ…ƒπŸ„ΈπŸ„½πŸ„Ά
❀‍πŸ”₯1πŸ‘€1🫑1
DoomPosting
Claude Code doesn't show you how many tokens you're using for subscriptions. No breakdown by model. No breakdown by project. Just a progress bar that says "63% used." So I built a local dashboard that reads the files Claude Code already writes to your machine.…
Alright so that confirms it?

This guy spent ~$200 for ~5M uncached, and ~2B cached tokens on Claude

I spent under $200 for $17M uncached and $6B cached on Codex

= Claude has at least 3x worse limits than Codex $20 plans,

Probably more like 10x worse, will do the math more exactly in a bit

πŸ„³πŸ„ΎπŸ„ΎπŸ„ΌπŸ„ΏπŸ„ΎπŸ…‚πŸ…ƒπŸ„ΈπŸ„½πŸ„Ά
πŸ’―1πŸ‘€1
This media is not supported in your browser
VIEW IN TELEGRAM
Guns aren’t even the most efficient way to influence a government, bombs are. The IRA’s vehicle bomb attacks was more effective than any gun attack. They’d smuggle in precursor chemicals from abroad and patiently build their bombs before targeting government and financial centers.

They’d give a telephoned pre-warning to authorities so civilians would be evacuated. With each bombing the economic impact was getting higher and higher. Insurance and taxes were increased, this made the government very unpopular with the people.

πŸ„³πŸ„ΎπŸ„ΎπŸ„ΌπŸ„ΏπŸ€–πŸ…‚πŸ…ƒπŸ„ΈπŸ„½πŸ„Ά
⚑3πŸ”₯1πŸ‘€1
This is absolutely crazy.

Someone just turned $13,200 into $472,000 betting that US-Iran ceasefire would happen today.

Making a 350x/3,500% gain in a single trade.

Lucky trader or Trump admin Insider ?

πŸ„³πŸ„ΎπŸ„ΎπŸ„ΌπŸ„ΏπŸ€–πŸ…‚πŸ…ƒπŸ„ΈπŸ„½πŸ„Ά
FOLLOWUP (0.83, took 0.2s):

β€” NEW: The two-week ceasefire plan includes allowing both Iran and Oman to charge transit fees on ships passing through the Strait of Hormuz, according to AP, citing a regional official.

β€” Iran will use the revenue for reconstruction, while Oman’s plans for the funds remain unclear.

πŸ„³πŸ„ΎπŸ„ΎπŸ„ΌπŸ„ΏπŸ€–πŸ…‚πŸ…ƒπŸ„ΈπŸ„½πŸ„Ά
😁1