Vibes and Cryptography
United States
Joined March 2017
- Tweets8.3K
- Following1.4K
- Followers590
- Likes8.9K
💀 retweeted
hate to give LLMs any credit but it’s pretty funny that they all have a cadence that it automatically goes into if it detects that you’re stupid.
💀 retweeted
We have a few Mac minis left over from our Daytona event in San Francisco.
Thought we'd give 3 away to people building something cool.
To enter:
1. Follow @daytonaio
2. Reply with what you'd build with a Mac mini and tag 2 people who'd build something cool with one.
We'll randomly pick 3 winners and announce them on October 30th.
Winners will be contacted directly via DM.
Good luck!
No purchase necessary. US residents 18+. Rules below.
Codex gets a lot of shit but the Claude $200 usage is basically just as bad. 85% usage gone in 24 hours.
37
If your argument is wanting intelligence too cheap to meter just use Luna. Obviously if you're using the frontier model your usage will drain faster. Common sense ain't that common I guess.
59
Check out @Muse, your personal AI agent. Redeem my code in Settings within 48 hours of joining and we'll both get 1 billion Muse tokens.
Code: JNQPVL
muse.ai/join
73
💀 retweeted
looks like codex limits probably weren't nerfed after all.
@sdmat123 went through usage across two pro 20x accounts and the quota burn is remarkably consistent across resets.
his reasoning regarding complaints:
> astra is just a very large model and burns through a lot of tokens.
> any efficiency gains get eaten up by the higher per token cost.
> around launch, people had multiple banked resets to burn through
it makes sense now that I think of it.
most of us had those banked resets and once those dried up, the normal weekly quota started feeling much tighter.
that said, usage can still feel rough sometimes depending on your workload and usage patterns. but the underlying limits themselves appears to be same.
Results are in: OAI hasn't changed anything for Astra usage in Codex.
I analyzed quota and tokens across two 20x accounts on my main dev system, and since launch usage is somewhere north of 800M tokens per reset. Overwhelmingly cached input.
The minor inconsistency is due to some Luna tokens (per-model in comments) and some light use elsewhere.
Claimed the Bitcoin Movie Enigma. 100k sats escrow swept, 99,766 received.
The trap: 24 listed BIP39 words with a bad checksum. Oracles that filter first never even derived it. Title prefixes + 5 IMDb groups (drop 2 outliers each) recover the phrase; BIP84 m/84'/0'/0'/0/0 matches the payout key.
Issue #24 / PR #25 in floflo777/open-crypto-puzzles. Community film IDs from #9.
1
1
99
Thanks @cryptop1r4t3 for building a puzzle.
And thanks @0xFlorent_ for open-crypto-puzzles. The film notes, oracle, and #9 thread made this solvable instead of another abandoned hunt.
1
1
119
💀 retweeted
Excited to announce our new @firecrawl startup deal!
More companies are being founded than ever & we want to power every one of them.
Now startups get 2x their monthly credits for a full year (up to $30k), plus a Slack channel with our team and swag.
firecrawl.dev/startups
